Loss of CCR5 results in glucose intolerance in diet-induced obese mice

Abstract

Macrophage and T cell infiltration into metabolic tissues contributes to obesity-associated inflammation and insulin resistance (IR). C-C chemokine receptor 5 (CCR5), expressed on macrophages and T cells, plays a critical role in the recruitment and activation of proinflammatory M1 and T_H1 immune cells to tissues and is elevated in adipose tissue (AT) and liver of obese humans and mice. Thus, we hypothesized that deficiency of CCR5 would protect against diet-induced inflammation and IR. CCR5-deficient (CCR5^−/−) mice and C57BL/6 (WT) controls were fed 10% low-fat (LF) or 60% high-fat (HF) diets for 16 wk. HF feeding increased adiposity, blood glucose, and plasma insulin levels equally in both genotypes. Opposing our hypothesis, HF-fed CCR5^−/− mice were significantly more glucose intolerant than WT mice. In AT, there was a significant reduction in the M1-associated gene CD11c, whereas M2 associated genes were not different between genotypes. In addition, HF feeding caused a twofold increase in CD4⁺ T cells in the AT of CCR5^−/− compared with WT mice. In liver and muscle, no differences in immune cell infiltration or inflammatory cytokine expression were detected. However, in AT and muscle, there was a mild reduction in insulin-induced phosphorylation of AKT and IRβ in CCR5^−/− compared with WT mice. These findings suggest that whereas CCR5 plays a minor role in regulating immune cell infiltration and inflammation in metabolic tissues, deficiency of CCR5 impairs systemic glucose tolerance as well as AT and muscle insulin signaling.

low-grade chronic inflammation is a common characteristic of obesity and its associated metabolic disorders, such as type 2 diabetes and cardiovascular disease. The immune system has been identified as a major regulator of inflammation in metabolic tissues (20, 39, 44). During obesity, proinflammatory macrophages and T cells are recruited to adipose tissue (AT) and secret various inflammatory mediators such as TNFα and CCL2, which impair insulin signaling and lead to the development of insulin resistance (IR) (13, 33, 39, 42). This immune cell infiltration occurs not only in AT but also in other metabolic tissues, including liver and muscle (17, 28, 30, 32), resulting in increased systemic IR and glucose intolerance. Identification of key mediators responsible for leukocyte recruitment and activation in metabolic tissues remains an area of great interest for therapeutic treatment of obesity-associated inflammation and IR.

Chemokines stimulate chemotaxis of immune cells through binding to specific G protein-coupled receptors. The seminal studies by Weisberg et al. (40) and Xu et al. (44) showing an increase in macrophages in obese compared with lean AT utilized microarrays to demonstrate an increase in chemokine and chemokine receptor expression in obese AT. In fact, the observation that obese AT is more inflammatory than lean AT has been known for some time (10, 38). Furthermore, in rodents and humans, plasma levels and AT expression of chemokines and their receptors are positively correlated with IR (11, 45). The chemokine CCL2 and its receptor CCR2 are highly elevated in obese AT. Kanda et al. (13) demonstrated that overexpression of CCL2 increases recruitment of AT macrophages (ATMs), leading to IR. Our group and others have demonstrated that CCR2^−/− mice fed a high-fat (HF) diet for a prolonged period of time are partially protected against ATM infiltration, inflammation, and glucose intolerance (9, 19, 39). Similarly, CXCL5 and its receptor CXCR2 have been shown to reduce ATM content and improve insulin sensitivity in obese mice (8, 26, 36). Despite these published reports, much remains to be learned regarding the role of individual chemokines and their receptors in macrophage as well as T cell recruitment to AT in obesity.

Recently, T cells have also been identified as regulators of macrophage recruitment and polarization, thereby orchestrating AT homeostasis (27, 33). Proinflammatory T_H1 CD4⁺ and CD8⁺ T cell number are elevated in obese human and rodent AT. In mice, C-C chemokine receptor 5 (CCR5) is expressed primarily on T_H1-polarized T cells, on NK cells, and, to a lesser extent, on monocytes (18, 21). This receptor is important for the recruitment and activation of T_H1 cells into sites of inflammation (24, 37). CCR5 binds and responds to the chemokines CCL3, CCL4, and CCL5 (31). In previous studies, we have investigated the impact of CCL3 deficiency on diet-induced obesity and found no protection against ATM and T cell infiltration or IR (14). However, because CCR5 binds multiple chemokines, targeting CCR5 may provide a direct role for this receptor on immune cell infiltration in obesity-induced IR.

CCR5 gene expression is elevated in the AT and liver of obese humans and mice (4, 11, 43). It is known that CCR5 deficiency reduces atherosclerosis in mice by reducing the number of macrophages in lesions (5). Thus we hypothesized that CCR5 deficiency would result in protection from the recruitment of inflammatory macrophages and T cells to AT in obesity. During the testing of our hypothesis, a published report demonstrated that CCR5 deficiency protects against diet-induced IR, hepatic steatosis, and AT inflammation by impairing M1 macrophage activation (16). These findings suggested that CCR5 is a major regulator of macrophage recruitment and activation in AT. However, upon analysis of the inflammatory and metabolic phenotype of CCR5^−/− mice in our laboratory, we found that CCR5 deficiency only mildly impacted macrophage polarization and T cell recruitment to metabolic tissue and actually led to systemic glucose intolerance under obese conditions.

MATERIALS AND METHODS

Mice.

All animal care procedures were performed in accordance with the guidelines of and after approval from the Vanderbilt University Institutional Animal Care and Use Committee. C57BL/6 [wild-type (WT)] and CCR5^−/− mice on a C57BL/6 background were originally purchased from Jackson Laboratories (Bar Harbor, ME). The WT and CCR5^−/− mice were crossed and the F1 CCR5^+/− mice used to generate WT and CCR5^−/− littermates for body weight, plasma glucose, and insulin (Fig. 1) and ATM quantification (Fig. 2). All other experiments were conducted using WT mice from our mouse colony (founders originally purchased from Jackson Laboratories) and CCR5^−/− mice bred from the F2 generation offspring, as described above. At 10 wk of age, mice were ad libitum fed a low-fat (LF; 10% fat, 3.58 kcal/g; D12450B) or HF diet (60% fat, 5.2 kcal/g; D12492) (Research Diets, New Brunswick, NJ). At the end of the study, mice were fasted for 5 h, bled via the retroorbital venous plexus, perfused at physiological pressure with PBS, and tissues were collected. Blood was centrifuged and plasma collected and frozen in aliquots until use.

Fig. 1.

C-C chemokine receptor 5 (CCR5) deficiency leads to impaired glucose tolerance during high-fat (HF) diet feeding. Wild-type (WT) and CCR5^−/− mice were maintained on a low-fat (LF) or HF diet for 16 wk. A: body weight curves (n = 10 LF WT; n = 8 LF CCR5^−/−; n = 22 HF WT; n = 23 HF CCR5^−/−). At 16 wk mice were fasted for 5 h, and blood collected. B and C: serum insulin (B) and blood glucose (C) were measured (n = 8–11 mice/group). D: after 16 wk of HF feeding, mice were fasted for 5 h, and basal blood glucose levels were measured (0 min) before intraperitoneal administration of a bolus of glucose according to lean body mass (1.0 g/kg). Blood glucose was assessed at 15, 30, 45, 60, 90, and 150 min after injection (n = 12–13 mice/group). Area under the curve was calculated using GraphPad Prism. Data are presented as means ± SE of 22–26 mice/group. Data in B–D were analyzed by 2-way ANOVA. Area under the curve in D was analyzed by Student's t-test. *P < 0.05.

Fig. 2.

CCR5 deficiency reduces M1 gene expression in adipose tissue (AT) during HF diet feeding. Following 16 wk of LF or HF diet, stromal vascular cells were collected from AT and analyzed by flow cytometry. A: representative flow plots of isolated AT stromal vascular cells gated for macrophages and lymphocytes (P1). Cells were further separated by DAPI staining for live and dead cells. Representative flow plot of live cells gated for AT macrophages identified by expression of CD11b⁺F4/80⁺ (n = 3 LF WT; n = 3 LF CCR5^−/−; n = 16 HF WT; n = 14 HF CCR5^−/−) and quantification of AT macrophages. Data in A were analyzed by 2-way ANOVA. B: AT was collected from WT and CCR5^−/− mice. Tissues were stained with an antibody to F4/80 (green) for macrophages and DAPI (blue) for nuclei. Tissues were analyzed by confocal microscopy using a ×40 objective. Sections were chosen from images representing mice with F4/80 gene expression close to the mean of their respective groups. C: AT was collected from WT and CCR5^−/− mice maintained on a HF diet for 16 wk. RNA was isolated and used for real-time RT-PCR analysis, as described in materials and methods. Data are the mean ± SE of the relative gene expression for 12 mice/group. Data in C were analyzed by Student's t-test. *P < 0.05.

Glucose tolerance test.

Mice were fasted for 5 h, and basal blood glucose levels were measured (0 min) before intraperitoneal administration of 1.0 g dextrose/kg lean body mass. Blood glucose was assessed at 15, 30, 45, 60, 90, and 150 min after injection.

Body composition analysis.

Mice were analyzed for total body fat and lean body mass at baseline and every 2 wk for a total of 16 wk. These analyses were performed by nuclear magnetic resonance, using the Bruker Minispec (Woodlands, TX) at the Vanderbilt University Mouse Metabolic Phenotyping Center.

Immunofluorescence analysis of AT.

Perigonadal AT was fixed in 1% paraformaldehyde. After fixation, tissue was washed with PBS and blocked with 10% goat serum. To stain for macrophages, a 1:100 dilution of anti-rat F4/80 (Abcam, Cambridge, MA) in goat serum was applied overnight at 4°C. Tissue was then washed with PBS and stained with a 1:1,000 dilution secondary antibody conjugated to anti-rat Alexa 488 (Cell Signaling Technology, Boston, MA). As a nuclear stain, 1 μg/ml DAPI was applied and tissue mounted on a glass slide with 90% glycerol. Images were acquired with a FV-1000 confocal microscope.

RNA isolation and real-time RT-PCR.

RNA was isolated from ∼100 mg of perigonadal AT using the RNeasy mini kit from Qiagen (Valencia, CA). cDNA was synthesized using the iScript cDNA synthesis kit from Bio-Rad (Hercules, CA). cDNA was diluted 1:2 or 1:10 and then used for real-time RT-PCR analysis on a Bio-Rad iQ5 machine. Primer/probe sets were purchased from the “Assays-on-Demand” program at Applied Biosystems (Foster City, CA). Quantification of 18S was performed for each sample, and final relative concentration was determined by comparing each gene of interest to 18S using the ΔΔC_T method.

AT stromal vascular fraction separation.

AT stromal vascular separation was performed as described (29). Briefly, epididymal AT pads from the mice were excised and minced in 3 ml of 0.5% BSA in PBS. After mincing, 2 mg/ml collagenase was added to the minced fat and incubated at 37°C for 20 min with shaking. The cell suspension was filtered through a 100-μm filter and then spun at 300 rpm for 10 min to separate floating adipocytes from the stromal vascular fraction pellet. The stromal vascular fraction pellet was suspended in ACK lysis buffer and incubated at room temperature for 5 min to lyse red blood cells. Cells were washed and centrifuged twice with PBS. Finally, cells were resuspended in FACS buffer (PBS, 1% FBS, 2 mM EDTA) and stained for flow cytometry.

Immune cell isolation from liver.

Liver was excised and minced in 1 mg/ml collagenase in PBS. Minced liver was incubated at 37°C for 30 min. The cell suspension was filtered through a 100-μm filter and spun at 300 rpm for 3 min. The supernatant was collected and centrifuged at 1,500 rpm for 10 min. The pellet was resuspended in 40% Percoll and overlayed on top of 60% Percoll. The Percoll gradient was centrifuged at 2,000 rpm for 20 min. The two middle layers of the Percoll gradient were collected in 3% FBS in RPMI and centrifuged at 1,500 rpm for 10 min. The supernatant was discarded and the pellet resuspended in FACS buffer and stained for flow cytometry.

Flow cytometry.

Cells isolated from various tissues were first incubated with Fc block for 5 min at room temperature, followed by incubation for 20 min at 4°C with fluorophore-conjugated antibodies F4/80-APC (1:100; eBioscience), CD11b-APC-Cy7 (1:200; BD Bioscience), CD4-Alex700 (1:100; BD Bioscience), TCRβ-APC (1:200; BD Bioscience), and CD8-PE-Cy7 (1:200; BD Bioscience). Samples were processed on a 5-Laser LSRII machine in the Vanderbilt Flow Cytometry Core and data analyzed using FlowJo software.

Western analysis.

Western blots were performed as described previously (1). Briefly, protein concentrations were determined using the Pierce BCA Protein Assay (Rockford, IL). Membranes were blocked for 1 h in Odyssey blocking buffer (LI-COR, Lincoln, NE) at room temperature. Membranes were then probed with specific antibodies for phosphorylated Akt Ser⁴⁷³ (p-Akt), total Akt (t-Akt), and β-actin (Cell Signaling Technology). Blots were visualized using the Odyssey Infrared Imaging System (LI-COR, Lincoln, NE). For the Odyssey, membranes were incubated with goat anti-rabbit IRDye 800CW secondary antibody at 1:10,000 dilution with 0.1% Tween-20 in PBS for 60 min, protected from light. After washing in PBS + 0.1% Tween-20, the membranes were scanned using the Odyssey Infrared Imaging System. Band intensity was quantified using Image J64 software.

Immunoprecipitation.

Immunoprecipitations of insulin signaling proteins were performed on AT and muscle homogenates, as described previously (41). Briefly, mice were fasted for 5 h and injected intraperitoneally with insulin (0.5 U/kg body wt) or saline. Fifteen minutes after injection, epididymal fat pads and muscle were collected and frozen in liquid nitrogen. Adipose tissue samples were homogenized in buffer (25 mM Tris·HCl, pH 7.4, 10% glycerol, 10 mM EDTA, 1% Triton X-100, 50 mM sodium pyrophosphate, 100 mM sodium fluoride, 1 mM PMSF, and protease inhibitors) and particulates removed by centrifugation. The supernatants were incubated overnight at 4°C with antibodies against insulin receptor substrate (IRS-1) or IRβ (Cell Signaling Technology), followed by the addition of protein A/G-Plus-Agarose (Santa Cruz Biotechnology, Santa Cruz, CA) for 2 h at 4°C. Immunoprecipitated proteins were resolved by SDS-PAGE. Membranes were immunoblotted for IRβ, IRS-1, or pTyr (PY99) (Santa Cruz Biotechnology), and signals were detected by chemiluminescence or the Odyssey system.

Statistical analysis.

Comparisons between recipients of WT and CCR5^−/− mice on the HF diet were performed using unpaired Student t-tests. Comparisons to detect diet and genotype effects were performed using two-way ANOVA.

RESULTS

CCR5 deficiency worsens HF diet-induced glucose intolerance.

To examine the impact of CCR5 deficiency on diet-induced obesity and IR, WT and CCR5^−/− mice were placed on a LF or HF diet for 16 wk. HF feeding significantly increased body weight, fat mass, and lean mass in WT and CCR5^−/− mice; however, no differences were detected between genotypes (Fig. 1A and Table 1). Similarly, HF feeding increased epididymal AT and liver mass as well as liver triglyceride (TG) content and plasma cholesterol in WT and CCR5^−/− mice (Table 1). No differences in plasma TG (Table 1), muscle TG, or food intake (data not shown) were detected between groups. There was a significant increase in plasma insulin (P < 0.05) and blood glucose levels (P < 0.05) in HF-fed mice compared with LF-fed controls of both genotypes (Fig. 1, B and C). After 8 wk of HF feeding, no difference in glucose tolerance was detected between groups (data not shown). However, following 16 wk of HF feeding, CCR5^−/− mice had an unexpected impairment in glucose tolerance compared with WT mice (P < 0.05; Fig. 1D).

Table 1.

Tissue masses and lipid levels after 16 wk of LF or HF diet

	Total Fat Mass, g	Epididymal Fat Mass, g	Lean Tissue Mass, g	Liver Mass, g	Liver TG, μg/mg	Serum Cholesterol, mg/dl	Serum TG, mg/dl
LF
WT (n = 10)	3.89 ± 1.78	0.68 ± 0.26	20.54 ± 1.44	1.37 ± 0.29	13.32 ± 1.30	75.68 ± 3.80	47.14 ± 1.19
CCR5−/− (n = 8)	2.71 ± 0.74	0.45 ± 0.14	19.61 ± 2.04	1.20 ± 0.23	14.33 ± 5.09	59.65 ± 9.32	47.33 ± 2.86
HF
WT (n = 36)	17.32 ± 3.22*	2.13 ± 0.50*	24.19 ± 3.23*	2.31 ± 0.72**	113.50 ± 10.90**	130.30 ± 7.86**	51.21 ± 2.54
CCR5−/− (n = 39)	17.13 ± 3.61*	2.10 ± 0.51*	23.72 ± 2.57*	2.16 ± 0.58**	105.00 ± 8.67**	127.30 ± 9.05**	45.25 ± 2.32

Data are the mean ± SE of the no. of mice in each group. LF, low fat; HF, high fat; TG, triglyceride; WT, wild type; CCR5, C-C chemokine receptor 5. WT and CCR5^−/− mice were placed on a LF or HF diet for 16 wk.

^*P < 0.05;

^**P < 0.001, diet effect for comparison with respective LF groups

CCR5 deficiency reduces M1 macrophage expression and increases CD4⁺ T cells in obese AT.

Based on the strong association between obesity and impaired glucose tolerance with macrophage infiltration into AT, we investigated whether CCR5 deficiency regulated ATM number and/or inflammatory status. As expected, there was a significant increase in AT CD11b⁺F4/80⁺ macrophages in mice fed HF diets (Fig. 2A); however, no difference between genotypes was detected. Similarly, immunofluorescence staining of F4/80⁺ cells in AT revealed no readily apparent difference in infiltration or localization between WT and CCR5^−/− mice on the HF diet (Fig. 2B). Interestingly, with regard to AT inflammatory status, the M1 marker CD11c was reduced significantly (P < 0.05) in CCR5^−/− mice compared with WT mice on the HF diet. A trend toward a reduction in TNFα (P < 0.07) was also detected in the AT of CCR5^−/− mice. However, no differences in M2 markers were discerned among the two genotypes (Fig. 2C). CD4⁺ T_H1 cells have recently been identified as key regulators of AT inflammation and IR (27, 33). Because CCR5 is expressed predominantly on T_H1-polarized cells and plays a critical role in their recruitment and activation in inflammation, we investigated the extent to which CCR5 deficiency regulated AT T cell populations during HF diet feeding. Flow cytometry analysis (Fig. 3A) revealed a surprising and significant elevation of CD4⁺ T cells in AT of obese CCR5^−/− mice compared with WT mice (P < 0.05; Fig. 3, B and C). In contrast, CD8⁺ T cell populations were not altered by CCR5 deficiency (Fig. 3C). Because CD4⁺ T cells can have a T_H1 or T_H2 phenotype, gene expression of T_H1/T_H2 markers (CCL4, CCL5, IFNγ, IL-2, and IL-12) in total AT was examined; however, there were no differences between groups (Fig. 3D).

Fig. 3.

CCR5 deficiency increases CD4⁺ T cells in AT during HF diet feeding. WT and CCR5^−/− mice were maintained on a HF diet for 16 wk. Flow cytometry was used to quantify CD4⁺ and CD8⁺ T cells in AT. A: representative flow plots of isolated AT stromal vascular fraction (SVF) gated for live cells (left) and then the live cells gated for TCRβ⁺ cells (right). B: representative flow plots of AT CD4⁺ and CD8⁺ T cells identified by expressing TCRβ⁺CD4⁺ and TCRβ⁺CD8⁺. C: quantification of AT CD4⁺ and CD8⁺ T cells. Data are the mean ± SE of 12 mice/group. Data in B were analyzed by Student's t-test. D: RNA from isolated AT was used for real-time RT-PCR analysis, as described in materials and methods. Data are the mean ± SE of the relative gene expression for 12 mice/group. Data in C and D were analyzed by Student's t-test. *P < 0.05.

CCR5 deficiency slightly impairs insulin signaling in AT and muscle of obese mice.

Although under obese conditions the CCR5^−/− mice had a mild reduction in M1 macrophage markers in AT (Fig. 2C), they were significantly more glucose intolerant compared with WT mice (Fig. 1D). Thus, the AT inflammation did not explain the systemic metabolic phenotype. As expected due to the IR in HF-fed mice, insulin induced only a small response in the phosphorylation of insulin-signaling proteins Akt, IRβ, and IRS-1 in AT of WT mice (Fig. 4, A–C). Insulin-induced phosphorylation of Akt and IRβ was blunted in CCR5^−/− compared with WT mice. Obesity impacts the inflammatory status of not only AT but also metabolic tissues such as liver and muscle. Analysis of liver immune cells (Fig. 5A) revealed no difference in macrophage populations (Fig. 5B) or CD4⁺ and CD8⁺ T cell populations (Fig. 5C) between HF-fed WT and CCR5^−/− mice. Likewise, no differences in inflammatory cytokine expression between WT and CCR5^−/− mice were detected in liver (Fig. 5D) or muscle (Fig. 6A). Interestingly, no difference in p-Akt was detected in liver (data not shown). In muscle, there was a mild reduction in phosphorylation of Akt in HF-fed CCR5^−/− mice injected with insulin compared with WT mice (Fig. 6B). However, no differences in insulin-stimulated phosphorylation of tyrosine of IRβ or IRS-1 were detected between genotypes (Fig. 6, C and D).

Fig. 4.

CCR5 deficiency mildly impairs phosphorylation of Akt in AT during obesity. At 16 wk of HF diet, WT and CCR5^−/− mice were injected with saline or insulin (0.5 U/kg) 15 min prior to euthanization. A: protein levels of p-Akt Ser⁴⁷³, β-actin, and total Akt (t-Akt) were determined by Western blot. Blots were analyzed densitometrically to quantify p-Akt relative to t-Akt. β-Actin was used as a loading control. Data represent the mean ± SE of 3–4 mice/roup. Equal amounts of AT protein lysates were immunoprecipitated (IP) with anti-IRβ (B) or IRS-1 antibody (C), followed by Western blot analyses with anti-p-Tyr, anti-IRβ, or anti-IRS-1, as indicated. Data represent the mean ± SE of 3 mice/group. Data in A–C were analyzed by 1-way ANOVA. Open bars represent WT mice, and black bars represent CCR5^−/− mice. IB, immunoblot.

Fig. 5.

CCR5 deficiency does not impact liver immune cell infiltration or inflammation under obese conditions. WT and CCR5^−/− mice were maintained on a HF diet for 16 wk. Flow cytometry was used to quantify macrophages and T cells in liver. A: representative flow plot of isolated hepatic nonparenchymal (NP) cells gated for macrophages and lymphocytes (P1). Cells were further separated by DAPI staining for live and dead cells. B: representative flow plot of hepatic macrophages identified by expression of CD11b⁺F4/80⁺ and quantification. Data are the mean ± SE of 3–5 mice/group. C: representative flow plot and quantification of liver CD4⁺ and CD8⁺ T cells identified by expressing TCRβ⁺CD4⁺ and TCRβ⁺CD8⁺. Data are the mean ± SE of 12 mice/group. D: RNA was isolated and used for real-time RT-PCR analysis, as described in materials and methods. Data are the mean ± SE of the relative gene expression for 12 mice/group. Data in B–D were analyzed by Student's t-test. SSC, side scatter; FSC, forward scatter.

Fig. 6.

CCR5 deficiency mildly impairs phosphorylation of Akt in muscle during obesity. Muscle was collected from WT and CCR5^−/− mice maintained on a HF diet for 16 wk. A: RNA was isolated and used for real-time RT-PCR analysis, as described in materials and methods. Data are the mean ± SE of the relative gene expression for 10–11 mice/group. B–D: after 16 wk of HF diet, WT and CCR5^−/− mice were injected with saline or insulin (0.5 U/kg) for 15 min prior to euthanization. B: protein levels of p-Akt Ser⁴⁷³, β-actin, and t-AKT were determined by Western blot. Blots were analyzed densitometrically to quantify p-Akt relative to t-Akt. β-Actin was used as a loading control. Data represent the mean ± SE of 3–4 mice/group. Equal amounts of muscle protein lysates were immunoprecipitated with anti-insulin receptor-β (IRβ; C) or IRS-1 (D) antibody, followed by Western blot analyses with anti-p-Tyr, anti-IRβ, or anti-IRS-1. Data represent the mean ± SE of 3 mice/group. Data in A were analyzed by Student's t-test. Data in B–D were analyzed by 1-way ANOVA.

DISCUSSION

The chemokine receptor CCR5 is classically defined as a regulator of macrophage and T cell recruitment and activation. We used CCR5^−/− mice to investigate the impact of CCR5 deficiency on immune cell recruitment to metabolic tissues during the development of obesity-induced inflammation and IR. Surprisingly, systemic glucose homeostasis was not improved by CCR5 deficiency. In fact, it was further impaired in the CCR5^−/− mice compared with WT controls. This phenotype appeared to be independent of inflammation, as deletion of CCR5 had a minimal impact on inflammatory immune cell populations in metabolic tissues. We found a significant reduction in CD11c and an increase in CD4⁺ T cells in AT of the CCR5^−/− mice. However, this did not coincide with any changes in levels of inflammatory cytokines. Likewise, we detected no differences in inflammation in liver or muscle. However, we found a mild reduction in Akt and IRβ phosphorylation in the AT and muscle of HF-fed CCR5^−/− mice. Thus, our findings demonstrate that under obese conditions, CCR5 has a minimal role in regulating immune cell activation and inflammation in metabolic tissue, and loss of this receptor does not protect against impaired glucose tolerance.

The worsened systemic glucose tolerance in CCR5^−/− mice was unexpected, especially given a recent report demonstrating improved glucose tolerance and insulin sensitivity in these animals (16). Systemic glucose tolerance is impacted primarily by two mechanisms: insulin secretion and tissue glucose uptake. Our original hypothesis was that CCR5 deficiency would reduce inflammation in metabolic tissues, leading to increased glucose uptake relative to WT mice and thereby improving glucose tolerance. Instead, our data point to a possible impairment of insulin signaling in AT and muscle of obese CCR5^−/− mice. This did not correlate with an elevation in muscle inflammation, suggesting that other factors may be responsible for this impairment. Although the average body weight and lean body mass were not different between WT and CCR5^−/− mice, we noticed an interesting difference in glucose tolerance between the genotypes, depending on the lean body mass of the mice. Because the dose of glucose was based upon lean body mass, no correlation is expected between lean body mass and glucose tolerance (area under the curve of the GTT). Accordingly, no such correlation was observed in the WT mice (Fig. 7A). In contrast, there was a significant positive correlation between lean body mass and glucose tolerance in CCR5^−/− mice (Fig. 7B). This led us to analyze the GTT data after separating both genotypes of mice into those with lean body mass less than or greater than 23 g. Interestingly, although this separation did not change the GTT results of WT mice (Fig. 7C), the impairment in glucose tolerance in CCR5^−/− mice was detected only in mice with lean body mass greater than 23 g (Fig. 7D). These data suggest the possibility of an inability of the pancreas to compensate for increased body mass by increasing insulin secretion in the heavier CCR5^−/− mice. Further studies will be necessary to determine whether CCR5 impacts β-cell function.

Fig. 7.

CCR5 deficiency impairment of glucose tolerance is positively correlated with lean body mass. Regression analysis of lean body mass vs. area under the curve (AUC) calculated from glucose tolerance test (GTT) in WT (n = 22; A) and CCR5^−/− mice (n = 26; B). Linear regression was calculated using GraphPad Prism. GTTs of WT and CCR5^−/− mice after 16 wk of HF feeding were separated based on lean body mass <23 g (C) and lean body mass >23 g (D) (n = 11–18 mice/group).

In a recent report, Kitade et al. (16) described their findings regarding inflammation and metabolism in CCR5^−/− mice. Similarly to our studies, they showed a decrease in AT expression of the M1 marker CD11c. Despite this similarity, most other findings were different between our groups. For example, Kitade et al. (16) used flow cytometry to assess M2 macrophages in AT and showed an increase in M2 ATMs in the CCR5^−/− mice. In contrast, our real-time RT-PCR analysis of M2 markers in whole AT demonstrated no difference between the groups. In addition, our data regarding systemic glucose tolerance showed completely opposite findings. We show that CCR5 deficiency impaired glucose tolerance slightly, whereas the previous study found improvements in glucose tolerance. One explanation for this is the different conditions used to perform the GTTs. In the study by Kitade et al. (16), the fasting period before the overnight GTT was compared with 5 h in our study. Prolonged fasting can have a profound impact on the metabolic profile of mice (2, 22). During overnight fasting conditions, mice are in a catabolic state and can lose ∼15% of their lean body mass. As mentioned above, we observed a significant association between lean body mass and glucose intolerance in HF-fed CCR5^−/− mice. Therefore, it is possible that overnight fasting may improve the outcome of GTT preferentially in CCR5^−/− compared with WT mice. An interesting point regarding our GTT studies was that the major impairment in glucose tolerance was noted before the 30-min time point, after which the slopes of the lines between the WT and CCR5^−/− mice were similar. This suggests that the increased glucose intolerance may be the result of impaired insulin secretion rather than impaired tissue insulin responsiveness. This concept is supported by the studies of Kitade et al. (16), where in their GTTs they see less insulin secretion, and in their insulin tolerance tests they see no difference in the rise in glucose after the 60-min time point in the CCR5^−/− mice. This explanation is also supported by the decrease in plasma insulin levels of HF-fed CCR5^−/− mice in their study. These vast differences in metabolic outcomes between the findings of Kitade et al. (16) and our findings highlight the potential confounding factors comparing and drawing conclusions from metabolic analysis in studies performed by different groups.

The well-established role for chemokines and chemokine receptors in other metabolic diseases, such as atherosclerosis (3, 6, 7), led naturally to the hypothesis that chemokines would be major players in the recruitment of macrophages to AT, leading to increased inflammation and reduced insulin sensitivity. Indeed, many different groups have assessed multiple chemokines and chemokine receptors for their role in AT inflammation and IR (34). The first groups to study a role of CCL2 in macrophage recruitment to AT showed a decrease in ATMs in CCL2^−/− mice (13). However, subsequent groups were not able to reproduce this finding (12, 15). With regard to the receptor for CCL2, CCR2, most studies have shown a reduction in ATMs (9, 19, 39). Strikingly though, differences in ATM numbers, inflammation, and insulin sensitivity are not detected until after long periods of HF diet feeding despite a nearly complete absence of circulating inflammatory monocytes. Deficiency of CCL3 (MIP-1α) was shown by our group to have no effect on ATM number (14, 35). Similarly, deficiency of the fractalkine receptor CX3CR1 has no effect on ATMs or IR (23). CXCL14, a chemokine targeting activated macrophages, was found to reduce ATM infiltration and inflammation and improve insulin sensitivity in obese mice (25). However, knockout of this chemokine in genetic and diet-induced obese mice causes lower body weight and food intake (36), suggesting that deletion of CXCL14 may have a more profound impact on regulation of body weight than directly on ATM infiltration and inflammation. Although a role for chemokines in AT inflammation is strongly supported by human and mouse data, use of single-gene knockout models to assess one chemokine at a time may be complicated by compensation of other chemokines. An alternative and intriguing possibility is that chemokine-independent mechanisms for immune cell accumulation in AT also exist.

In conclusion, our studies demonstrate that deficiency of CCR5 moderately reduces M1 macrophage infiltration but increases CD4⁺ T cell infiltration into AT during obesity. However, this did not lead to improvements in systemic glucose tolerance. Thus, targeting only CCR5 may not be the best approach for treating inflammation and IR associated with obesity.

GRANTS

This work was supported by a Career Development Award from the American Diabetes Association to A. H. Hasty (1-07-CD-10). A. J. Kennedy was also supported by an American Diabetes Association mentor-based postdoctoral fellowship (7-10-MI-05) and by a UNCF-Merck Postdoctoral Science Research Fellowship. A. A. Hill was supported by the Training in Cardiovascular Research Program (T32HL007411). Confocal microscopy was performed through the use of the Vanderbilt University Medical Center (VUMC) Cell Imaging Shared Resource (supported by National Institutes of Health Grants CA-68485, DK-20593, DK-58404, HD-15052, DK-59637, and EY-08126). Flow cytometry experiments were performed in the VUMC Flow Cytometry Shared Resource, which is supported by the Vanderbilt Digestive Disease Research Center (DK-058404). Lipid profiles were performed at the lipid core of the Mouse Metabolic Phenotyping Center at Vanderbilt University (DK-59637).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.J.K. and A.H.H. contributed to the conception and design of the research; A.J.K., C.D.W., A.A.H., M.L.G., and L.G.J. performed the experiments; A.J.K. and L.G.J. analyzed the data; A.J.K. and A.H.H. interpreted the results of the experiments; A.J.K. prepared the figures; A.J.K. and A.H.H. drafted the manuscript; A.J.K., C.D.W., A.A.H., M.L.G., L.G.J., and A.H.H. approved the final version of the manuscript; C.D.W., A.A.H., M.L.G., and A.H.H. edited and revised the manuscript.