Obesity is associated with low-grade inflammation, insulin resistance, type 2 diabetes, and cardiovascular disease. This study investigated the effect of a 15-wk lifestyle intervention (hypocaloric diet and daily exercise) on inflammatory markers in plasma, adipose tissue (AT), and skeletal muscle (SM) in 27 severely obese subjects (mean body mass index: 45.8 kg/m2). Plasma samples, subcutaneous abdominal AT biopsies, and vastus lateralis SM biopsies were obtained before and after the intervention and analyzed by ELISA and RT-PCR. The intervention reduced body weight (P < 0.001) and increased insulin sensitivity (homeostasis model assessment; P < 0.05). Plasma adiponectin (P < 0.001) increased, and C-reactive protein (P < 0.05), IL-6 (P < 0.01), IL-8 (P < 0.05), and monocyte chemoattractant protein-1 (P < 0.01) decreased. AT inflammation was reduced, determined from an increased mRNA expression of adiponectin (P < 0.001) and a decreased expression of macrophage-specific markers (CD14, CD68), IL-6, IL-8, and tumor necrosis factor-α (P < 0.01). After adjusting for macrophage infiltration in AT, only IL-6 mRNA was decreased (P < 0.05). Only very low levels of inflammatory markers were found in SM. The intervention had no effect on adiponectin receptor 1 and 2 mRNA in AT or SM. Thus hypocaloric diet and increased physical activity improved insulin sensitivity and reduced low-grade inflammation. Markers of inflammation were particularly reduced in AT, whereas SM does not contribute to this attenuation of whole body inflammation.
- weight loss
obesity, especially visceral obesity, and physical inactivity are strong predictors of morbidity and mortality (26). Visceral obesity is considered to be a chronic low-grade inflammatory state associated with insulin resistance, type 2 diabetes, and cardiovascular disease (13, 27). Human adipose tissue (AT) is characterized by the ability to produce and release inflammatory proteins, collectively known as adipokines, e.g., adiponectin, tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-8, and monocyte chemoattractant protein (MCP)-1 (38). Circulating levels and AT release of TNF-α, IL-6, IL-8, and MCP-1 are increased in obesity and reduced after weight loss (2, 3, 11, 14). Visceral AT (VAT) seems to be more closely associated with the inflammatory state than subcutaneous AT (SAT) since higher amounts of IL-6 (20), IL-8 (5), and MCP-1 (6) are released from the VAT depot. In contrast to other adipokines, adiponectin seems to have anti-inflammatory properties and is found to be decreased in obesity (1), increased after weight loss (25), and expressed in lower amounts in VAT compared with SAT (30). Not only AT but also skeletal muscle (SM) produces and releases several cytokines (29), and in SM especially IL-6 has attracted attention since both mRNA expression and release of IL-6 from the muscle tissue are highly upregulated in response to acute exercise (34). By itself, moderate long-term exercise training seems to have no effect on the circulating levels of proinflammatory cytokines (31). However, combined with a hypocaloric diet, moderate exercise training reduces inflammation more than diet alone (44).
The relationship between (visceral) obesity, low-grade inflammation, and insulin resistance is complex, but accumulating evidence indicates a possible interaction between human AT and SM (37). One of the possible links between the AT and SM might be the adipose-specific protein adiponectin, which is found in high concentrations in the circulation of lean subjects (1). Interestingly, low levels of adiponectin are associated with increased morbidity (25) and high levels of C-reactive protein (CRP) and IL-6 (7, 17). In vitro, adiponectin decreases pro-inflammatory properties of TNF-α (33) and increases anti-inflammatory cytokines such as IL-10 and IL-1 receptor antibody (41). In vivo, treatment of rodents with adiponectin increases insulin sensitivity in SM (21, 43) and liver (43), probably through activation of the AMP-activated protein kinase system via adiponectin receptor type 1 (adipoR1) and 2 (adipoR2) in SM (adipoR1) and liver (adipoR2; see Ref. 42). Thus low adiponectin levels in obesity seem to be associated with increased inflammation and morbidity.
Because activation of the inflammatory system seems to be an integrated pathway through which obesity is leading to health complications, we wanted in the present study to investigate how conventional lifestyle intervention consisting of a hypocaloric diet and daily moderate exercise would affect markers of inflammation (CRP, TNF-α, IL-6, IL-8, MCP-1, and adiponectin) both in the circulation and in the AT from severely obese subjects. Furthermore, because SM recently has been suggested to be an endocrine organ producing and secreting numerous cytokines, the effect of this lifestyle intervention on the expression of these cytokines was also investigated in SM.
MATERIALS AND METHODS
The participants were recruited from a center (Ebeltoft Kurcenter) established to treat severe obesity. In an open study, 27 (15 females and 12 males) severely obese subjects [mean body mass index (BMI): 45.8 ± 1.7 kg/m2, BMI range: 31.4–63.0 kg/m2] were consecutively included in a 15-wk lifestyle intervention program consisting of a hypocaloric diet and moderate daily physical activity. The hypocaloric diet was calculated to reduce the subject's body weight by ∼1%/wk according to the individual gender, age, body weight, and level of physical activity. The individual exercise program consisted of at least 2–3 h of moderate-intensity physical activity (e.g., walking, swimming, aerobics) 5 days/wk. The diet was calculated by a dietician and evaluated one time per week, and a physiotherapist supervised the physical activity. Five of the subjects were treated for hypertension (with calcium channel blockers, ANG I converting enzyme inhibitors, or ANG II receptor blockers), four were treated for asthma (only inhalation), and five received oral contraceptives, but all received the same dose and type of medication throughout the study. Body composition was assessed by bioelectrical impedance (Quantum X; RJL Systems), blood-pressure was assessed by a fully automatic blood pressure monitoring system (Omron IC; Hutchings Healthcare, West Sussex, UK), and the waist circumference was measured at the midpoint between the upper iliac crest and lower thoracic rib at the level of the umbilicus.
After the 15-wk intervention period, 23 subjects (12 females and 11 males) with a mean BMI of 45.8 ± 1.9 kg/m2 (range: 31.4–63.0 kg/m2) remained in study. Four subjects (3 females and 1 male) left the center before the 2nd day of the experiment and were therefore excluded from the study. Experiments were performed before (baseline) and repeated after the 15-wk intervention period. At least 6 h before the time of the experiments, the subjects were fasting and had not engaged in vigorous physical activity, and fasting and inactivity periods were standardized between pre- and postintervention experiments. The participants went through a general examination, including anthropometrical assessments, before having blood samples taken. Next, AT and SM biopsies were performed followed by an oral glucose tolerance test (OGTT). Finally, maximal aerobic capacity was assessed through an indirect symptom-limited graded exercise test, where submaximal oxygen uptake was measured on-line (Sensor Medics; Spiropharma) at three consecutive 5-min work loads. Plasma samples for determination of fasting insulin and fasting glucose were analyzed at the local clinical biochemical laboratory, and a measure of insulin resistance was obtained using the homeostasis model assessment (HOMA = fasting insulin × fasting glucose/22.5; see Ref. 22). AT samples were obtained from the subcutaneous, abdominal AT depot as previously described (9) at baseline and after the 15 wk in 19 subjects (10 females and 9 males; mean BMI of 47.4 ± 2.0 kg/m2). The skin was anesthetized with lidocaine (10 mg/ml) before a small incision was made, and ∼200 mg of AT were removed under sterile conditions. Immediately after removal, the AT sample was washed in isotonic NaCl, snap-frozen in liquid nitrogen, and kept at −80°C until RNA extraction. SM samples were obtained from the vastus lateralis muscle before and after the 15 wk in 14 subjects (7 females and 7 males) with a mean BMI of 48.0 ± 2.1 kg/m2 (range: 36.3–63.0 kg/m2). Skin and muscle fascia were anaesthetized with lidocaine (5 mg/ml), and under sterile conditions a 1-cm incision was made, whereafter ∼100 mg of muscle tissue were removed using the Bergström technique. Immediately after removal, the muscle biopsy was dissected free of visible fat, snap-frozen in liquid nitrogen, and kept at −80°C until RNA extraction.
Determination of inflammatory markers in plasma.
Plasma levels of adiponectin, IL-6, IL-8, TNF-α, and MCP-1 were measured using specific high-sensitive human ELISA. Plasma levels of CRP were measured using an immunoturbidimetric assay. The adiponectin assay (B-Bridge International) had an intra-assay coefficient of variation (CV) of 5.0% (n = 12). The IL-8 assay (Quantiglo; R&D Systems Europe, Abingdon, UK) had an intra-assay CV of 6.4% (n = 12). The IL-6 assay (Quantikine HS600; R&D Systems Europe) had an intra-assay CV of 5.5% (n = 12). The TNF-α assay (Quantikine HSTA00C; R&D Systems Europe) had an intra-assay CV of 3.5% (n = 12). The MCP-1 assay (R&D Systems Europe) had an intra-assay CV of 8.1% (n = 12). The CRP assay (Roche Diagnostics, Mannheim, Germany) had an intra-assay CV of 1.8% (n = 20).
Determination of mRNA levels.
The oligonucleotide primer pairs used for the mRNA determination are listed in Table 1. RNA was isolated using TRIzol reagent (GIBCO-BRL Life Technologies, Roskilde, Denmark), and cDNA was made with random hexamer primers using the GeneAmp PCR kit (Applied Biosystems). The mRNA levels of the target genes (adiponectin, adipoR1, adipoR2, IL-6, IL-8, MCP-1, TNF-α, CD68, or CD14) were expressed relative to the housekeeping gene (β-actin), except in Fig. 2B, where IL-6, IL-8, MCP-1, and TNF-α were target genes and CD68 was used as a housekeeping gene to evaluate the impact of resident macrophages. The expression of β-actin was unchanged before and after the intervention (threshold cycles: 25.3 ± 0.3 vs. 24.9 ± 0.3; P = 0.29). Quantification was performed with a SYBR Green real-time PCR assay using an iCycler PCR machine (Bio-Rad Laboratories, Hercules, CA), as previously described (8). In brief, PCR amplification was performed with PCR mastermix containing target primers, Hot Star Taq DNA polymerase, SYBR Green, and PCR buffer. All samples were determined as duplicate. Samples were incubated for an initial denaturation at 95°C for 10 min, followed by 40 PCR cycles each consisting of 95°C for 30 s, 57°C for 30 s, and 74°C for 60 s. Relative gene expression of target gene to β-actin was calculated as described in User Bulletin No. 2 from PerkinElmer (PerkinElmer Cetus, Norwalk, CT).
The SigmaStat 3.1 statistical packet (Systat Software, Point Richmond, CA) was used for the calculations. Normality was tested using the Kolmogorov-Smirnov test. Paired t-test was used for comparison between anthropometrical data, body composition, parameters of insulin sensitivity, circulating levels of various adipokines, and mRNA levels in AT and SM biopsies before and after weight loss. To determine the relationship between plasma levels of adipokines, metabolic or anthropometrical parameters, and adipokine mRNA levels at baseline and after weight loss, a bivariate correlation with a Pearson correlation coefficient (rp) was used. Values are presented as means ± SE. Threshold for significance was set at P < 0.05.
Informed, written consent was obtained from all subjects, and experiments were performed in accordance with the Helsinki II Declaration. The study was approved by the Ethical Committee of Copenhagen (KF 01-220/03).
Effects of weight loss on anthropometric and metabolic parameters.
The obese subjects reduced their body weight by ∼18 kg (138.3 ± 5.9 vs. 120.7 ± 5.4 kg; P < 0.001) during the intervention. There was no gender difference in total weight loss or percentage of weight loss (female vs. male: 17.2 vs. 18.0 kg and 14.0 vs. 13.2%). Moreover, reduction was observed in relation to BMI (45.8 ± 1.9 vs. 40.0 ± 1.8 kg/m2; P < 0.001), waist circumference (142.6 ± 4.0 vs. 132.2 ± 4.2 kg/m2; P < 0.001), and total body fat mass (46.0 ± 2.5 vs. 41.4 ± 2.3 kg; P < 0.001). Compared with baseline, insulin sensitivity was improved since area under the OGTT curve was significantly reduced after the intervention (P < 0.001, Fig. 1). As a marker of insulin resistance, HOMA was decreased significantly after the 15-wk intervention (2.1 ± 0.3 vs. 1.3 ± 0.2; P < 0.05). Compared with baseline, no significant difference was observed in blood pressure (BP) after the intervention (systolic BP: 131.1 ± 2.7 vs. 128.6 ± 2.7 mmHg; diastolic BP: 83.8 ± 2.3 vs. 79.2 ± 3.0 mmHg). Finally, a significant 29% increase in the maximal aerobic capacity was observed after the intervention (21.7 ± 1.4 vs. 28.2 ± 2.2 ml O2·min−1·kg−1; P < 0.001).
Effect of weight loss on plasma levels of inflammatory markers.
As shown in Table 2, the 15-wk intervention resulted in a significant decrease in circulating levels of CRP (P < 0.05), IL-6 (P < 0.01), IL-8 (P < 0.05), and MCP-1 (P < 0.01) as well as a significant increase in circulating levels of adiponectin (P < 0.001). The intervention had no effect on circulating levels of TNF-α (Table 2).
Effects of weight loss on adipokine mRNA levels in AT.
AT mRNA levels of the macrophage-specific markers CD68 and CD14 were decreased by ∼40% (P < 0.05, Fig. 2A) and ∼55% (P < 0.01, Fig. 2B), respectively, after the intervention. No gender difference was observed in relation to the changes in macrophage-specific markers. As expected, adiponectin mRNA significantly increased (4.5 ± 0.8 vs. 6.3 ± 0.9 arbitrary units; P < 0.001) after the intervention. In the AT biopsies, mRNA levels of IL-6, IL-8, and TNF-α, but not MCP-1, were significantly reduced after the 15-wk intervention (P < 0.05, Fig. 3A). Interestingly, when adjusting for the amount of macrophages by adjusting for CD68 mRNA expression in the AT biopsies, only IL-6 mRNA was found independently to be significantly decreased after the intervention (P < 0.05, Fig. 3B). Adjusting for macrophage infiltration had, however, no effect on the enhanced adiponectin expression observed after the intervention (P < 0.01, Fig. 3B). The AT mRNA levels of adipoR1 were ∼10-fold higher than mRNA levels of adipoR2, but no difference was observed in mRNA levels of the two receptors before and after the 15-wk intervention (Table 3). At baseline, AT mRNA levels of the macrophage-specific markers CD68 and CD14 were significantly correlated to BMI (rp = 0.61; p < 0.01), and positive correlations to the adipokine expression were also found: TNF-α mRNA (rp = 0.58; P < 0.01), IL-6 mRNA (rp = 0.80; P < 0.001), and IL-8 mRNA (rp = 0.47; P < 0.05).
Effects of weight loss on adipokine mRNA levels in the SM.
The mRNA levels of the macrophage-specific markers CD68 and CD14 were found to be very low in the SM samples, and no differences in CD68 and CD14 mRNA were observed after the 15-wk intervention (Fig. 2, C and D), suggesting that the mRNA signal of the two macrophage-specific markers may be because of contamination. In the SM biopsies, mRNA levels of IL-6 but not IL-8 and TNF-α decreased after the 15-wk intervention (Fig. 4). The SM mRNA levels of adipoR1 were approximately fivefold higher than SM mRNA levels of adipoR2, but the intervention had no effect on the mRNA levels of the two receptors (Table 3). No correlation was observed between BMI and any of the investigated adipokines or the macrophage-specific markers in the SM samples.
Low-grade inflammation seems to be a common finding in obesity, type 2 diabetes, and cardiovascular disease (13). In obesity, increased levels of inflammatory proteins in the circulation may partly reflect a spillover from the AT (23, 38), and several studies have demonstrated that weight loss reduces low-grade inflammation in AT (2, 7) and plasma (11, 25, 44).
The present study investigated the effects of a 15-wk lifestyle intervention consisting of a hypocaloric diet and daily moderate physical activity in severely obese subjects. The novel finding in the present study was the lifestyle intervention-induced reduction in inflammation in plasma and AT, paralleled by a significant improvement in metabolic status. In the circulation, CRP, IL-6, IL-8, and MCP-1 were reduced and adiponectin increased. In AT biopsies, similar findings were obtained, with significant reduction in general levels of inflammation as determined by changes in adipokine expression (decrease in IL-6, IL-8, and TNF-α and increase in adiponectin) and a decrease in expression of macrophage-specific markers (CD68 and CD14). Interestingly, after adjusting for AT resident macrophages by using the expression of the macrophage-specific marker CD68 as housekeeping gene, only IL-6 expression was significantly independently decreased. It suggests that the changes in expression of TNF-α and IL-8 primarily may be attributable to changes in the number of AT resident macrophages and that weight loss is associated with reduced infiltration of macrophages in the AT. In SM biopsies, only IL-6 expression was significantly reduced after the intervention. SM expression of CD68 and CD14 was very low and not affected by the 15-wk intervention, suggesting that, in contrast to the findings in the AT, none or only minor infiltration of macrophages is present in SM of severely obese subjects. This is in line with recent findings by Di Gregorio et al. (16).
Except for leptin and adiponectin, the majority of adipokines are produced and released from nonadipose or stromal vascular (SV) cells resident in the AT (6, 19). The SV fraction consists of ∼10% of CD14+ macrophages (12), and macrophage infiltration in human SAT (40) and VAT (6) has been reported to be increased in a linear relationship with increasing adiposity (e.g., BMI). Weight loss obtained through hypocaloric diet, exercise, or a combination of both is characterized by a reduction in overall adiposity (loss of SAT and VAT) accompanied by an attenuation of inflammation in the SAT depot [reduction in protein levels of TNF-α (24), IL-6 (2, 3), and IL-8 (2) and increase in protein levels of adiponectin (7)]. In contrast, a reduction in SAT mass alone, obtained through abdominal liposuction, does not lead to a similar decrease in low-grade inflammation (28), indicating that the VAT depot is more closely associated with the inflammatory state in obesity than the SAT depot. Recently, surgery-induced weight loss (gastric bypass) in severely obese subjects was demonstrated to reduce macrophage infiltration in SAT (10). However, the present paper demonstrates for the first time that weight loss obtained through lifestyle intervention induces a pronounced reduction in macrophage infiltration in SAT paralleled by a reduction in inflammation in the circulation (reduced CRP, TNF-α, IL-6, and IL-8 and increased adiponectin). In contrast to expression of TNF-α and IL-8 in AT, the decrease in IL-6 expression does not seem to be attributable only to the decrease in macrophage infiltration but also to differences in their primary origin (adipocytes vs. macrophages vs. endothelial cells). TNF-α is primarily produced in the SV fraction of the AT, whereas IL-6 and IL-8 are produced and released to a comparable extent (∼15–20%) from isolated human adipocytes compared with SV cells (6, 18). Recently, human endothelial cells were reported to release higher levels of IL-6 than IL-8 (4), suggesting that endothelial cells within the AT may be responsible for some of the difference between IL-6 and IL-8 expression after correcting for macrophage infiltration. In addition, differences in responsiveness to metabolic changes associated with weight loss and physical activity or other yet unidentified proteins/pathways may be involved in the regulation of the production of the two adipokines.
SM has the ability to release several inflammatory proteins (e.g., TNF-α, IL-6, IL-8, and IL-1β) upon induction by either an exogenous (lipopolysaccharide; see Ref. 29) or an endogenous (vigorous exercise-induced muscle damage; see Refs. 32 and 39) stimulus. Short-term muscle contraction in the absence of obvious muscle damage induces the release of significant amounts of IL-6 but not TNF-α (36), and moderate long-term exercise either alone (35) or combined with a hypocaloric diet (44) increases insulin sensitivity and decreases circulating levels of several inflammatory markers. Because vigorous exercise has been demonstrated to give rise to macrophage infiltration in SM (39), the very low mRNA levels of the macrophage-specific markers CD68 and CD14 found in the present study may indicate that daily moderate physical activity is not associated with macrophage infiltration and thereby inflammation in SM. SM-derived IL-6 increases robustly in relation to short-term muscle contraction or vigorous exercise, but in our study we found that weight loss associated with moderate physical activity decreased IL-6 expression in SM. Thus muscle activation may exhibit a biphasic IL-6 response in relation to both intensity and duration of exercise; however, the mechanism behind this remains to be elucidated.
The anti-inflammatory adipokine, adiponectin, has been reported to exert its effects through activation of two distinct receptors (42). In the present study, mRNA levels of adipoR1 were higher than mRNA levels of adipoR2 in both AT and SM. The findings presented are essentially in line with the original report where adipoR1 was higher in SM, although more ubiquitously expressed than adipoR2, which seems to be restricted to the liver (42). Even though the 15-wk intervention in the present study resulted in a significant increase in adiponectin levels in both AT and circulation paralleled by an increase in insulin sensitivity, mRNA levels of adipoR1 and adipoR2 remained unchanged in both AT and SM. The regulation of the two receptors awaits further investigations since their regulation may include changes in other pathways than changes in adiponectin levels per se.
In conclusion, the combination of hypocaloric diet and moderate physical activity resulted in a significant general decrease in the level of inflammation in plasma and AT, paralleled by a significant improvement in metabolic parameters in severely obese subjects. Of specific interest is that low adiponectin levels in AT and plasma of the severely obese subjects seems not to be related to the changes (increase) in AT macrophage infiltration but may rather be the result of inhibitory effects of adipokines (TNF-α and IL-6) released from the macrophages resident in the AT (7). Overall, the observed reduction in low-grade inflammation seems mainly to be related to a reduction in macrophage infiltration in the AT, whereas SM probably to only a minor degree is involved in the observed changes in inflammatory markers.
The study was supported by the Novo Nordic Foundation, Konsul J. Fogh-Nielsens Legat, and The Danish Medical Research Council (22-02-0527).
We thank subjects and staff at Ebeltoft Kurcenter for participating in the study. We highly appreciate the expert technical assistance of Lenette Pedersen, Pia Hornbek, and Regitze Kraunsøe. Finally, we thank Jane Østergaard Pedersen and Esther Zimmermann for performing the anthropometrical measurements and assisting the subjects on the days of the experiments.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by American Physiological Society