Role of CC Chemokine Receptor 2 in Bone Marrow Cells in the Recruitment of Macrophages into Obese Adipose Tissue*

The MCP-1 (monocyte chemoattractant protein-1)/CCR2 (CC motif chemokine receptor-2) pathway may play a role in macrophage infiltration into obese adipose tissue. Here we investigated the role of CCR2 in the recruitment of bone marrow-derived macrophages into obese adipose tissue in vitro and in vivo. Using the TAXIScan device, which can measure quantitatively the directionality and velocity of cell migration at time lapse intervals in vitro, we demonstrated that bone marrow cells (BMCs) from wild type mice migrate directly toward MCP-1 or culture medium conditioned by adipose tissue explants of genetically obese ob/ob mice, which are efficiently suppressed by pharmacological blockade of CCR2 signaling. The number of F4/80-positive macrophages was reduced in the adipose tissue from high fat diet-fed obese KKAy or ob/ob mice treated with a CCR2 antagonist propagermanium relative to vehicle-treated groups. We also found that the number of macrophages is reduced in the adipose tissue from ob/ob mice reconstituted with CCR2-/- BMCs (ob/ob + CCR2-/- BMCs) relative to those with CCR2+/+ BMCs (ob/ob + CCR2+/+ BMCs). Expression of mRNAs for CD11c and TLR4 (Toll-like receptor 4) markers of proinflammatory M1 macrophages was also decreased in the adipose tissue from ob/ob + CCR2-/- BMCs relative to ob/ob + CCR2+/+ BMCs, whereas mannose receptor and CD163, markers of anti-inflammatory M2 macrophages, were unchanged. This study provides in vivo and in vitro evidence that CCR2 in bone marrow cells plays an important role in the recruitment of macrophages into obese adipose tissue.

The MCP-1 (monocyte chemoattractant protein-1)/CCR2 (CC motif chemokine receptor-2) pathway may play a role in macrophage infiltration into obese adipose tissue. Here we investigated the role of CCR2 in the recruitment of bone marrow-derived macrophages into obese adipose tissue in vitro and in vivo. Using the TAXIScan device, which can measure quantitatively the directionality and velocity of cell migration at time lapse intervals in vitro, we demonstrated that bone marrow cells (BMCs) from wild type mice migrate directly toward MCP-1 or culture medium conditioned by adipose tissue explants of genetically obese ob/ob mice, which are efficiently suppressed by pharmacological blockade of CCR2 signaling. The number of F4/80-positive macrophages was reduced in the adipose tissue from high fat diet-fed obese KKA y or ob/ob mice treated with a CCR2 antagonist propagermanium relative to vehicle-treated groups. We also found that the number of macrophages is reduced in the adipose tissue from ob/ob mice reconstituted with CCR2 ؊/؊ BMCs (ob/ob ؉ CCR2 ؊/؊ BMCs) relative to those with CCR2 ؉/؉ BMCs (ob/ob ؉ CCR2 ؉/؉ BMCs). Expression of mRNAs for CD11c and TLR4 (Toll-like receptor 4) markers of proinflammatory M1 macrophages was also decreased in the adipose tissue from ob/ob ؉ CCR2 ؊/؊ BMCs relative to ob/ob ؉ CCR2 ؉/؉ BMCs, whereas mannose receptor and CD163, markers of anti-inflammatory M2 macrophages, were unchanged. This study provides in vivo and in vitro evidence that CCR2 in bone marrow cells plays an important role in the recruitment of macrophages into obese adipose tissue.
Obesity is a state of chronic, low grade inflammation; it is characterized by increased infiltration of macrophages, suggesting that macrophages are an important source of inflammation in obese adipose tissue (1)(2)(3). Using an in vitro co-culture system composed of adipocytes and macrophages, we have demonstrated that a paracrine loop involving saturated fatty acids and tumor necrosis factor-␣ derived from adipocytes and macrophages, respectively, establishes a vicious cycle that augments the inflammatory changes (i.e. marked up-regulation of proinflammatory adipocytokines, such as MCP-1 (monocyte chemoattractant protein-1) and tumor necrosis factor-␣, and down-regulation of adiponectin) (4,5). These findings have led us to speculate that macrophages, when infiltrated, may participate in the inflammatory pathways that are activated in obese adipose tissue.
CCR2 (CC motif chemokine receptor-2) is a major receptor that recognizes MCP-1 and some other chemokines, which control the recruitment of macrophages to the site of inflammation (6,7). There is considerable evidence for the pathophysiologic role of the MCP-1/CCR2 pathway in macrophage infiltration into obese adipose tissue. For instance, Weisberg et al. (8) reported the attenuation of macrophage accumulation and inflammatory changes in the adipose tissue from mice lacking CCR-2 (CCR2 Ϫ/Ϫ mice) during a high fat diet. Moreover, two previous studies with transgenic mice overexpressing MCP-1 in the adipose tissue and/or MCP-1-deficient mice (MCP-1 Ϫ/Ϫ mice) showed that MCP-1 plays a role in the recruitment of macrophages into obese adipose tissue (9,10). Recent studies demonstrated the heterogeneity of macrophages in obese adipose tissue; they follow two different polarization states: M1 or "classically activated" (proinflammatory) macrophages and M2 or "alternatively activated" (anti-inflammatory) macrophages (11,12). Evidence has accumulated indicating an increase in M1 macrophages and decrease in M2 macrophages in obese adipose tissue (11)(12)(13). Interestingly, such a phenotypic switch was not observed in CCR2 Ϫ/Ϫ mice (11), suggesting that the MCP-1/CCR2 pathway contributes to macrophage polarization in the adipose tissue. Although macrophages infiltrated into obese adipose tissue have been reported to be mostly derived from the bone marrow (1), there is no direct evidence for the contribution of CCR2 to the recruitment of bone marrow-derived macrophages into obese adipose tissue.
In this study, we examined the role of CCR2 in the recruitment of bone marrow-derived macrophages into obese adipose tissue through a combination of an optically accessible, horizontal chemotaxis apparatus with which to measure cell migration at time lapse intervals in vitro and bone marrow transplantation techniques in vivo. The data from this study provide evidence that CCR2 in bone marrow cells (BMCs) 4 is involved in the recruitment of macrophages into obese adipose tissue.
Animal Studies-Generation of male CCR2 Ϫ/Ϫ mice on the C57BL/6J background was reported previously (15). Fourweek-old female C57BL/6J mice and ob/ob mice were purchased from Charles River Japan (Tsukuba, Japan). Four-weekold male KK mice and KKA y mice were purchased from Nippon CLEA (Shizuoka, Japan). The green fluorescence protein (GFP) transgenic mice were kindly provided by Dr. M. Okabe (Osaka University, Suita, Japan) (16). The animals were housed in individual cages in a temperature-, humidity-, and light-controlled room (12-h light and 12-h dark cycle) and allowed free access to water and standard chow (Oriental MF; 362 kcal/100 g, 5.4% energy as fat; Oriental Yeast) except when otherwise noted. At the end of the experiments, mice were sacrificed after a 1-h fast under intraperitoneal pentobarbital anesthesia (30 mg/kg). All animal experiments were conducted according to the guidelines of the Tokyo Medical and Dental University Committee on Animal Research (No. 0050140).
Administration of Propagermanium-Five-week-old male KKA y mice fed high fat diet were given a CCR2 antagonist propagermanium (8 mg/kg/day) (14) in drinking water for 4 weeks. Five-week-old female ob/ob mice were fed standard chow supplemented with propagermanium (3 or 8 mg/kg/day) for 5 weeks.
Bone Marrow Transplantation-BMCs were collected from the femur and tibia of female CCR2 ϩ/ϩ and CCR2 Ϫ/Ϫ donor mice. Red blood cells were depleted by ACK lysing buffer (0.15 M NH 4 Cl, 1 mM KHCO 3 , 0.1 mM EDTA), and the remaining cells were washed with PBS and resuspended at a density of 10 7 cells/ml in PBS containing 3% fetal bovine serum (Sanko Junyaku). BMCs (3 ϫ 10 6 cells) obtained from CCR2 ϩ/ϩ or CCR2 Ϫ/Ϫ donor mice were washed three times with cold PBS, and injected intravenously into 7 gray irradiated 7-week-old female wild type or ob/ob recipient mice. Each recipient mouse received CCR2 ϩ/ϩ or CCR2 Ϫ/Ϫ BMCs. The efficiency of bone marrow transplantation was determined by counting GFP-positive cells in peripheral blood of wild type or ob/ob mice reconstituted with BMCs obtained from GFP transgenic donor mice (16). In this study, more than 90% of peripheral blood cells were substituted by GFP-positive cells (data not shown). Eleven weeks after the bone marrow transplantation, the perigonadal adipose tissue was harvested and analyzed.
Histological Analysis-The perigonadal adipose tissue was fixed with neutral buffered formalin and embedded in paraffin. An immunohistochemical study was carried out using 5-m-thick paraffin-embedded sections for the macrophage marker F4/80 as previously described (5,17,18). The number of F4/80-positive cells was counted in a blinded fashion through the microscope with a ϫ400 objective and an eyepiece grid indicating a 0.0625-mm 2 field of view. More than 100 serial fields were examined, and the data were expressed as the mean number/mm 2 .
Blood Parameter Analysis-Blood was sampled from the tail vein of mice. Blood glucose was measured as previously described (19).
Preparation of Conditioned Medium-The perigonadal adipose tissue was harvested aseptically from 14-week-old female C57BL/6J wild type mice and ob/ob mice as previously described (20,21). Half of the adipose tissue was minced and then cultured for 24 h in Dulbecco's modified Eagle's medium containing 1% bovine serum albumin. The supernatants thus obtained were used for chemotaxis assay. The culture medium was used as a control.
Cell Culture-The CCR2 expression vector was constructed by ligating the full-length mouse CCR2 cDNA into the XhoI-NotI site of BCMGSNeo vector and transfected into Jurkat cells using FuGENE-HD transfection reagent (Roche Applied Science) according to the manufacturer's instructions. The Jurkat cells stably expressing CCR2 (CCR2-Jurkat) were obtained by selecting with 1 mg/ml G418 (Invitrogen). CCR2-Jurkat was maintained in RPMI (Nacalai Tesque) containing 10% fetal bovine serum (Biowest, Nuaille, France) and antibiotics and incubated at 37°C in a humidified 5% CO 2 , 95% air atmosphere. BMCs were collected from the femur and tibia, suspended to 10 6 cells/ml in RPMI 1640 medium supplemented with 20 mM HEPES buffer, pH 7.2, and kept at 37°C before use.
Real Time Horizontal Chemotaxis Assay Using TAXIScan-Chemotaxis of CCR2-Jurkat and BMCs was examined using the TAXIScan device (ECI, Inc., Tokyo, Japan) (22,23). CCR2-Jurkat was injected into one of two compartments through a hole connected to the compartment. To adjust the position of the cells, the medium was drawn out from the other hole provided for the other compartment. Drawing was stopped when cells were aligned in the channel close to the edge. One microliter of chemoattractant solution was injected into the other compartment to initiate chemotaxis and monitoring. Cell migration at 37°C was recorded every 1 min for 30 min, and the migration pathway was recorded by clicking each cell on a display. The directionality of migration is expressed as the angle (radian) toward the concentration gradient (i.e. /2 indicates that the cell is migrating toward the concentration gradient, whereas radian Ϫ/2 indicates that the cell is migrating against the gradient). The velocity of cell migration is expressed as m/s. The average directionality and velocity of migration were calculated by the values obtained every 1 min ( Fig. S1) (23,24). In this study, we confirmed that CCR2-Jurkat migrate dose-dependently toward recombinant mouse MCP-1; 100 nM was most effective to induce chemotaxis ( Fig. S2), which is consistent with a previous report (25). Thus, 100 nM recombinant MCP-1 was used for further analysis.
Quantitative Real Time PCR-Quantitative real time PCR was performed with an ABI Prism 7000 Sequence Detection System using TAQMAN or SYBR Green PCR Master Mix reagent kit (Applied Biosystems, Foster City, CA). Primers used were described in supplemental Table 1. Levels of mRNA were normalized to those of housekeeping gene 36B4 mRNA.
Enzyme-linked Immunosorbent Assay-The MCP-1 concentrations in culture medium conditioned by adipose tissue explants were determined by the commercially available enzyme-linked immunosorbent assay kits (R&D Systems, Minneapolis, NM).
Fluorescence-activated Cell Sorting Analysis-BMCs were harvested from the femur and tibia of CCR2 ϩ/ϩ or CCR2 Ϫ/Ϫ mice, and mononuclear cells were separated by HISTOPAQUE (Sigma). Cells were incubated with Fc receptor-blocking reagent (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were then stained in staining buffer (PBS containing 0.1% bovine serum albumin and 5 mM EDTA) using fluorochrome-conjugated combinations of the following antibodies: fluorescein isothiocyanate-conjugated anti-mouse CD11b and phycoerythrin-conjugated antimouse CD11c (BD Biosciences). After washing twice with the staining buffer, flow cytometry was performed on EPICS ALTRA (Beckman Coulter, Fullerton, CA).
Statistical Analysis-Data were expressed as the mean Ϯ S.E. Statistical analysis was performed using analysis of variance followed by Scheffe's test. p Ͻ 0.05 was considered to be statistically significant.

RESULTS
Macrophage Infiltration in CCR2 ϩ/ϩ and CCR2 Ϫ/Ϫ Mice with High Fat Diet-induced Obesity-To investigate the role of CCR2 in macrophage infiltration into obese adipose tissue, we examined the response of CCR2 ϩ/ϩ and CCR2 Ϫ/Ϫ mice to a high fat diet. Although there was no significant change in body weight between the genotypes, the weight of perigonadal adipose tissue was even increased in CCR2 Ϫ/Ϫ mice relative to CCR2 ϩ/ϩ mice (Fig. 1, A and B). Histological examination revealed that the number of F4/80-positive cells in the adipose tissue is markedly reduced in CCR2 Ϫ/Ϫ mice relative to CCR2 ϩ/ϩ mice ( Fig. 1C; p Ͻ 0.01). We found that CCR2 Ϫ/Ϫ mice have significantly lower blood glucose levels than CCR2 ϩ/ϩ mice 20 weeks after the high fat diet feeding (280.25 Ϯ 29.09 mg/dl versus 193.33 Ϯ 15.62 mg/dl; p Ͻ 0.05). These observations are consistent with the previous report on reduced macrophage infiltration in CCR2 Ϫ/Ϫ mice fed a high fat diet during the course of obesity (8).
Chemotaxis of BMCs toward MCP-1 and Obese Adipose Tissue-conditioned Medium-It has been reported that macrophages in obese adipose tissue are mostly derived from the bone marrow (1), but there is no direct evidence for the involvement of CCR2 expressed in BMCs. To explore the role of CCR2 in the recruitment of bone marrow-derived macrophages into obese adipose tissue, using the TAXIScan device, we examined the real time horizontal chemotactic response of mouse BMCs toward MCP-1 and adipose tissue-conditioned medium in vitro. The BMCs derived from wild type mice randomly migrated with the vehicle or PBS containing 3% bovine serum albumin. By contrast, BMCs migrated directly toward 10 Ϫ7 M recombinant MCP-1, the dose of which was most effective to induce chemotaxis (Fig. S2). The directionality of migration was abrogated by pharmacological blockade of CCR2 signaling with a CCR2 antagonist, propagermanium ( Fig. 2A). We also examined the chemotactic response of BMCs toward the culture medium conditioned by adipose tissue explants from  DECEMBER 19, 2008 • VOLUME 283 • NUMBER 51 ob/ob mice (ob/ob-conditioned medium). The BMCs randomly migrated with the control culture medium. By contrast, BMCs migrated toward ob/ob-conditioned medium directly and quickly. Furthermore, the directionality of migration was significantly abrogated by preincubation with propagermanium ( Fig. 2B; p Ͻ 0.01). We also obtained essentially the same results with another CCR2 antagonist, RS504393 (Fig. 2C). These observations, taken together, suggest that BMCs migrate toward MCP-1 or ob/ob-conditioned medium via CCR2. The MCP-1 concentrations in ob/ob-conditioned medium were ϳ150 ng/ml (i.e. 20 nM), which are roughly comparable with the dose of recombinant MCP-1 used in vitro (100 nM) (data not shown). In this study, we found that the migration of BMCs is direct toward ob/ob-conditioned medium relative to wild type-conditioned medium, whereas the velocity of migration is not differ-ent between the conditioned media used (Fig. S3). In addition, MCP-1 concentrations in ob/ob-conditioned medium were increased relative to wild type-conditioned medium (93.96 Ϯ 8.71 ng/ml versus 150.75 Ϯ 5.02 ng/ml, p Ͻ 0.01, n ϭ 5). These observations are consistent with the concept that increased production of MCP-1 is involved in macrophage infiltration into obese adipose tissue (8 -10).

CCR2 and Macrophage Recruitment into Obese Adipose Tissue
Chemotaxis of CCR2-Jurkat toward MCP-1 and Obese Adipose Tissue-conditioned Medium-Because of the heterogeneity of BMCs, we next generated Jurkat cells stably expressing CCR2 (CCR2-Jurkat) to confirm that CCR2 is involved in cell migration toward MCP-1 or ob/ob-conditioned medium. In this study, CCR2 mRNA levels in CCR2-Jurkat were roughly comparable with those in BMCs from wild type mice (Fig. S4). The CCR2-Jurkat migrated directly toward recombinant MCP-1 at a dose of 10 Ϫ7 M, whereas they moved randomly with the vehicle alone. By contrast, the directionality of migration was abrogated by preincubation with propagermanium ( Fig. 3A and supplemental Movies 1-3, p Ͻ 0.05). The CCR2-Jurkat also migrated toward ob/ob-conditioned medium with the directionality of migration being significantly inhibited by propagermanium ( Fig.  3B and supplemental Movies 6 -8). The parent Jurkat, a human T lymphocyte cell line, expresses CXCR4 (CXC motif chemokine receptor 4) and binds SDF-1 with high affinity (26). We previously reported that the parent Jurkat migrates toward recombinant human SDF-1 through CXCR4 (22). However, it was unable to migrate toward recombinant mouse MCP-1 (data not shown). We also confirmed that the directionality and velocity of CCR2-Jurkat migration toward recombinant human SDF-1 are unaffected by preincubation with propagermanium ( Fig. S5 and supplemental Movies 4 and 5), which suggests the specificity of propagermanium as a CCR2 antagonist in the chemotaxis assay. In this study, chemotaxis of CCR2-Jurkat toward ob/ob-conditioned medium was also significantly inhibited by another CCR2 antagonist, RS504393 (p Ͻ 0.05) (Fig. 3C). These observations indicate that migration of CCR2-Jurkat in response to recombinant MCP-1 or ob/ob-conditioned medium is CCR2dependent.

Chemotaxis of BMCs toward Culture Medium Conditioned by Adipose Tissue Explants from Pioglitazone-treated ob/ob
Mice-Previous reports showed that peroxisome proliferatoractivated receptor ␥ activation suppresses adipose tissue expression of MCP-1 and thus improves insulin sensitivity (27,28). We also examined the chemotaxis of BMCs toward culture medium conditioned by adipose tissue explants from pioglitazone-and vehicle-treated ob/ob mice (ob/ob ϩ pio-and ob/ ob ϩ veh-conditioned medium, respectively). In this study, we confirmed that MCP-1 and tumor necrosis factor-␣ mRNA expression is decreased and adiponectin mRNA expression is increased in the adipose tissue by pioglitazone treatment, which is consistent with previous reports (27,28) (Fig. S6). The BMCs migrated directly toward ob/ob-conditioned medium, and the directionality of migration of BMCs was markedly abrogated with ob/ob ϩ pio-conditioned medium (Fig. S6). On the other hand, there was no significant change in the velocity of migration between the conditioned medium used. These and velocity (right) of migration were determined from digital time lapse movies. B, effect of propagermanium on chemotaxis of CCR2 ϩ/ϩ BMCs toward ob/ob-conditioned medium. The directionality (left) and velocity (right) of migration were determined from digital time lapse movies. The CCR2 ϩ/ϩ BMCs were incubated with 5 g/ml propagermanium 30 min before and during the assay. C, effect of RS504393 on chemotaxis of CCR2 ϩ/ϩ BMCs toward ob/ob-conditioned medium. The directionality (left) and velocity (right) of migration were determined from digital time lapse movies. CCR2 ϩ/ϩ BMCs were incubated with 5 g/ml RS504393 30 min before assay and during the assay. veh, vehicle; ct, control; PG, propagermanium; RS, RS504393. *, p Ͻ 0.05; **, p Ͻ 0.01; N.S., not significant. n ϭ 4.
observations support the concept that treatment with a peroxisome proliferator-activated receptor ␥ agonist, pioglitazone, reduces the potential to induce macrophage infiltration into obese adipose tissue (11,12).
Effect of Propagermanium on Macrophage Infiltration into Obese Adipose Tissue-To explore the involvement of CCR2 in macrophage infiltration into obese adipose tissue in vivo, we examined the effect of propagermanium (8 mg/kg/day) in KKA y mice fed a high fat diet. Although there was no significant difference in body weight and adipose tissue weight between the propagermanium-and the vehicle-treated groups, the number of F4/80-positive cells was significantly decreased in the propagermanium-treated KKA y mice relative to the vehicle-treated group (Fig. 4, A-C, p Ͻ 0.01). In this study, we also found that the number of F4/80-positive cells is significantly reduced in ob/ob mice treated with propagermanium (3 mg/kg/day or 8 mg/kg/day) relative to the vehicle-treated group (p Ͻ 0.05), although body weight and adipose tissue weight are unchanged (Fig. 5, A-C). These observations, taken together, indicate that pharmacological blockade of CCR2 effectively reduces macrophage infiltration into obese adipose tissue.
Role of CCR2 in the Recruitment of Bone Marrow-derived Macrophages into Obese Adipose Tissue-Using the bone marrow transplantation technique, we examined whether bone marrow-derived macrophages are infiltrated into obese adipose tissue through CCR2 in vivo. There was no significant difference in body weight and adipose tissue weight between ob/ob mice reconstituted with CCR2 Ϫ/Ϫ BMCs (ob/ob ϩ CCR2 Ϫ/Ϫ BMCs) and those reconstituted with CCR2 ϩ/ϩ BMCs (ob/ob ϩ CCR2 ϩ/ϩ BMCs) (Fig. 6, A and B). Histological analysis revealed that the number of F4/80-positive cells is significantly reduced in the adipose tissue from ob/ob ϩ CCR2 Ϫ/Ϫ BMCs relative to ob/ob ϩ CCR2 ϩ/ϩ BMCs (Fig. 6C, p Ͻ 0.01). Moreover, expression of CD11c and TLR4 (Toll-like receptor-4), markers of M1 macrophages, was significantly decreased in ob/ob ϩ CCR2 Ϫ/Ϫ BMCs relative to ob/ob ϩ CCR2 ϩ/ϩ BMCs ( Fig. 7; p Ͻ 0.01). In this study, there was no significant difference in expression of mannose receptor and CD163, markers of M2 macrophages, between the genotypes of BMCs (Fig. 7). We also observed no significant difference in the population of CD11b ϩ CD11c ϩ cells in BMCs between the genotypes (Fig.  S7), suggesting that CCR2 is not involved in CD11b ϩ CD11c ϩ cell differentiation in the bone marrow. These observations, taken together, suggest that CCR2 expressed in BMCs, especially in M1 macrophages, plays an important role in the recruitment of bone marrow-derived macrophages into obese adipose tissue.

DISCUSSION
There is considerable evidence that CCR2 and its major ligand MCP-1 are important for macrophage accumulation in obese adipose tissue (8 -10, 29). Weisberg et al. (1) previously reported that macrophages in obese adipose tissue are mostly derived from the bone marrow. However, whether CCR2 expressed in BMCs contributes to the recruitment of macrophages into obese adipose tissue has not been addressed. Furthermore, there has been no quantitative analysis of CCR2mediated migration of BMCs in response to soluble factor(s) released from the adipose tissue in vitro even with conventional methods such as Boyden's chamber (30) and under-agarose technique (31)(32)(33), which have only enabled end point analysis of cell migration with limited information. This study was designed to elucidate the role of CCR2 in the recruitment of bone marrow-derived macrophages into obese adipose tissue both in vivo and in vitro.
Using the TAXIScan device, an optically accessible, horizontal chemotaxis apparatus, we demonstrated that BMCs from wild type mice migrate toward recombinant MCP-1 or ob/obconditioned medium. The data were also confirmed using the CCR2-Jurkat. Moreover, the chemotactic responses are all blunted by treatment with propagermanium, which targets gly-  1-3). B, effect of propagermainum on chemotaxis of CCR2-Jurkat toward the culture medium conditioned by adipose tissue explants from ob/ob mice (ob/ob). The directionality (left) and velocity (right) were determined from digital time lapse movies (supplemental Movies 6 -8). The CCR2 ϩ/ϩ BMCs were incubated with 5 g/ml propagermanium 30 min before and during the assay. C, effect of RS504393 on chemotaxis of CCR2 ϩ/ϩ BMCs toward ob/ob-conditioned medium. The directionality (left) and velocity (right) of migration were determined from digital time lapse movies. CCR2 ϩ/ϩ BMCs were incubated with 5 g/ml RS504393 30 min before and during the assay. veh, vehicle; ct, control; PG, propagermanium; RS, RS504393. *, p Ͻ 0.05; **, p Ͻ 0.01; N.S., not significant. n ϭ 4. DECEMBER 19, 2008 • VOLUME 283 • NUMBER 51 cosylphosphatidylinositol-anchored proteins closely associated with CCR2 and thus inhibits the MCP-1-induced chemotactic migration of monocytes (14). These observations suggest the chemotactic response is CCR2-dependent. In this study, we demonstrated that pharmacologic blockade of CCR2 by propagermanium for 4 weeks effectively reduces macrophage accumulation in the adipose tissue from both high fat diet-fed KKA y mice and ob/ob mice, which is consistent with the previous report with INCB3344, another CCR2specific antagonist (8). These observations, taken together, suggest that obese adipose tissue is capable of releasing a large amount of chemokine(s) involved in the migration of BMCs via CCR2, and one such factor might be MCP-1. We also found that BMCs migrate directly toward ob/ob-conditioned medium relative to wild type-conditioned medium and that MCP-1 concentrations in ob/ob-conditioned medium were increased relative to wild type-conditioned medium. Furthermore, the directionality of migration of BMCs is markedly abrogated with ob/ob ϩ pio-conditioned medium relative to ob/ob ϩ veh-conditioned medium, suggesting that treatment with peroxisome proliferator-activated receptor ␥ agonist reduces the potential to induce macrophage infiltration into obese adipose tissue (11,12). Therefore, TAXIScan provides a unique in vitro experimental system with which to assess quantitatively the chemotactic activity of chemokines responsible for cell migration in vivo.

CCR2 and Macrophage Recruitment into Obese Adipose Tissue
The in vitro chemotaxis assay system applied herein enables us to quantify the directionality and velocity of cell migration, which cannot be assessed by conventional methods. The in vivo implication of the directionality and velocity observed in vitro has not been fully elucidated; however, Terashima et al. (25) recently reported that a cytoplasmic protein FROUNT, which is implicated in the directionality of CCR2-mediated monocyte chemotaxis in vitro, is indispensable to macrophage infiltration in a murine peritonitis model in vivo. Given that the role of macrophages in obese adipose tissue has recently attracted the interest of many researchers of metabolic diseases, this study has introduced a new approach toward the understanding of the molecular mechanism underlying macrophage infiltration into obese adipose tissue.
In this study, we demonstrated that macrophage infiltration is reduced in adipose tissue from ob/ob ϩ CCR2 Ϫ/Ϫ BMCs relative to ob/ob ϩ CCR2 ϩ/ϩ BMCs. The data of this study suggest  that CCR2 expressed in BMCs is involved in the recruitment of bone marrow-derived macrophages into obese adipose tissue in vivo. In this regard, Tsou et al. (34) recently demonstrated by intravenous infusion of ex vivo-labeled CCR2 Ϫ/Ϫ BMCs into WT recipient mice with thioglycollateinduced peritonitis that CCR2 and its ligands MCP-1 and MCP-3 are necessary for efficient monocyte recruitment from the systemic circulation to the site of acute inflammation. On the other hand, there are a couple of reports that CCR2 also plays a role in monocyte recruitment from the bone marrow to the systemic circulation (34,35). In this regard, it would be interesting to know how CCR2 expressed in bone marrow-derived monocytes contributes to the recruitment of monocytes from the bone marrow to the systemic circulation and that from the systemic circulation to obese adipose tissue.
Recent studies have pointed to the heterogeneity of macrophages infiltrated into obese adipose tissue (i.e. macrophages in obese adipose tissue have at least two different activation states, M1 or "classically activated" (proinflammatory) and M2 or "alternatively activated" (anti-inflammatory) macrophages) (11,36). In this study, we demonstrated that mRNA expression of markers for M1-polarized macrophages, CD11c and TLR4, is decreased in ob/ob ϩ CCR2 Ϫ/Ϫ BMCs relative to ob/ob ϩ CCR2 ϩ/ϩ BMCs. On the other hand, there was no significant difference in expression of markers for M2-polarized macrophages, mannose receptor and CD163, between the genotypes of BMCs. Importantly, there was no appreciable difference in the population of CD11b ϩ CD11c ϩ cells in BMCs between the genotypes, suggesting that CCR2 is not involved in CD11b ϩ CD11c ϩ cell differentiation in the bone marrow. Given that CCR2 expression is enriched in proinflammatory M1 macrophages (37), the data of this study support the concept that CCR2 expressed in BMCs plays an important role in the recruitment of bone marrow-derived M1 macrophages into obese adipose tissue (11,29). Determination of whether CCR2 is sufficient to the recruitment of M1 macrophages into obese adipose tissue must await further investigation.
In this study, we observed that the number of macrophages is significantly reduced in CCR2 Ϫ/Ϫ mice relative to CCR2 ϩ/ϩ mice during a high fat diet. This is consistent with a previous report using CCR2 Ϫ/Ϫ mice on the C57BL/6 background, which is identical to ours (8). However, Chen et al. (38) did not observe such a difference in macrophage infiltration into obese adipose tissue using CCR2 Ϫ/Ϫ mice on the DBA background. On the other hand, there are several reports supporting the concept that MCP-1 does not contribute to the recruitment of macrophages into obese adipose tissue (39,40). Given the mildly reduced macrophage infiltration by genetic as well as pharmacologic blockade of FIGURE 6. Phenotypic characterization of ob/ob mice reconstituted with CCR2 ؉/؉ or CCR2 ؊/؊ BMCs. A, growth curve of ob/ob mice reconstituted with CCR2 ϩ/ϩ BMCs (ob/ob ϩ CCR2 ϩ/ϩ BMCs) and ob/ob mice reconstituted with CCR2 Ϫ/Ϫ BMCs (ob/ob ϩ CCR2 Ϫ/Ϫ BMCs) after bone marrow transplantation. Open circle, ob/ob ϩ CCR2 ϩ/ϩ BMCs; closed square, ob/ob ϩ CCR2 Ϫ/Ϫ BMCs. B, weight of the subcutaneous (sub), perigonadal (gona), and mesenteric (mes) adipose tissue. C, F4/80 immunostaining (left) and the number of F4/80positive cells (right) in the perigonadal adipose tissue. Original magnification was ϫ200. Scale bars, 100 m. *, p Ͻ 0.05; **, p Ͻ 0.01; N.S., not significant. n ϭ 4 -6. FIGURE 7. Characterization of adipose tissue macrophages from ob/ob mice. Shown is expression of mRNAs for CD11c and TLR4 (M1 macrophage markers) and that for mannose receptor (MR) and CD163 (M2 macrophage markers) in the perigonadal adipose tissue from WT ϩ CCR2 ϩ/ϩ BMCs, WT ϩ CCR2 Ϫ/Ϫ BMCs, ob/ob ϩ CCR2 ϩ/ϩ BMCs, and ob/ob ϩ CCR2 Ϫ/Ϫ BMCs. *, p Ͻ 0.05; **, p Ͻ 0.01; N.S., not significant. n ϭ 4 -6. DECEMBER 19, 2008 • VOLUME 283 • NUMBER 51 CCR2 in vivo, there might be other soluble factors that compensate, when CCR2 is disrupted, for the recruitment of macrophages into obese adipose tissue. For instance, recent studies have suggested that osteopontin and CXCL14 (CXC motif chemokine ligand-14) play a role in obesity-induced macrophage infiltration into adipose tissue (41,42). Indeed, we found that the chemotactic response of CCR2-Jurkat and BMCs is suppressed significantly but not completely by propagermanium, which may be explained by other soluble factor(s) or chemokine(s) secreted from obese adipose tissue. In this study, we also found that concentrations of osteopontin as well as MCP-1 are increased in ob/ob-conditioned medium relative to wild type-conditioned medium. 5 Furthermore, we have recently observed that mRNA expression of osteopontin, CCL5 (CC motif chemokine ligand-5), and macrophage colony-stimulating factor, which are all known to recruit macrophages to the site of inflammation, is markedly up-regulated in the adipose tissue from CCR2 Ϫ/Ϫ mice relative to CCR2 ϩ/ϩ mice during the course of obesity. 6 It is also important to note that CCR2 is able to recognize other monocyte chemoattractants, such as MCP-2, MCP-3, MCP-4, and MCP-5 (6). In this regard, our in vitro assay system herein may help identify novel adipocyte-derived chemokine(s) or even therapeutic agents that target macrophage infiltration into obese adipose tissue.

CCR2 and Macrophage Recruitment into Obese Adipose Tissue
In conclusion, this study highlights the role of CCR2 expressed in BMCs in the recruitment of bone marrow-derived macrophages into obese adipose tissue, thereby facilitating the better understanding of the pathophysiologic role of CCR2 in obesity-related metabolic syndrome.