Endocrine

, Volume 45, Issue 3, pp 382–391

Circulating chemerin decreases in response to a combined strength and endurance training

Authors

    • Medicobiological UnitInternational Scientific Institute, National Sports Academy
  • Matthias Blüher
    • Department of MedicineUniversity of Leipzig
  • Anna Vekova
    • Medicobiological UnitInternational Scientific Institute, National Sports Academy
  • Iveta Bonova
    • Medicobiological UnitInternational Scientific Institute, National Sports Academy
  • Stanislav Tzvetkov
    • Medicobiological UnitInternational Scientific Institute, National Sports Academy
  • Dimo Kurktschiev
    • Robert Koch Medical Centre
  • Theodora Temelkova-Kurktschiev
    • Robert Koch Medical Centre
Original Article

DOI: 10.1007/s12020-013-0003-2

Cite this article as:
Stefanov, T., Blüher, M., Vekova, A. et al. Endocrine (2014) 45: 382. doi:10.1007/s12020-013-0003-2

Abstract

Chemerin is an adipokine that may mediate the link between obesity, inflammation, insulin resistance, type 2 diabetes mellitus, and cardiovascular disease. In this study, we examined the association between chemerin and various cardiometabolic risk factors in cross-sectional setting and tested the hypothesis that a 6-month combined exercise program decreases serum chemerin in overweight or obese, non-diabetic individuals. Serum chemerin concentration was measured in a cross-sectional analysis including 98 individuals with a wide range of age and body mass index (BMI). In addition, chemerin was measured in 79 sedentary, overweight or obese, non-diabetic individuals who completed a 6-month combined endurance and resistance exercise program (CEP, n = 51) or served as controls (C, n = 28). Chemerin was significantly associated with total cholesterol (p = 0.04), triglycerides (p < 0.001), fasting insulin (p < 0.001), homeostasis model assessment of insulin resistance (HOMA-IR, p < 0.001), systolic blood pressure (p = 0.04), highly sensitive C-reactive protein (p = 0.03), leucocytes count (p = 0.047), and leptin (p = 0.008) independently of age and BMI. In multiple regression analysis, chemerin was an independent determinant of HOMA-IR. As a result of the 6-month training program, serum chemerin decreased significantly in CEP group (−13.8 ± 13.2 ng/ml, p < 0.001). A significant association between the changes in chemerin and improved HOMA-IR were found even after adjustment for changes in waist circumference. Among non-diabetic individuals serum chemerin was associated with various cardiometabolic risk factors independently of BMI. In addition, the 6-month combined strength and endurance training program led to a significant reduction in circulating chemerin levels in overweight or obese individuals.

Keywords

AdipokinesChemerinCombined exerciseObesityMetabolic diseaseCardiovascular disease

Introduction

The prevalence of obesity, a multifactorial health threat caused by a complex interplay between genetic predisposition and the environment [1], is constantly growing posing enormous health burden on societies worldwide [2]. Ectopic fat accumulation in visceral depots has been associated with a number of cardiometabolic disorders such as insulin resistance, fatty liver disease, the metabolic syndrome (MetS), type 2 diabetes mellitus (T2DM), and cardiovascular disease (CVD) [3]. It is now widely accepted that besides being an important energy depot, adipose tissue is a metabolically active organ that may regulate a variety of systemic processes through the secretion of soluble proteins, collectively known as adipokines [4]. Serum adipokine concentrations are closely related to the degree of obesity and altered adipokine secretion is hypothesized to mediate the relationship between obesity, inflammation, insulin resistance, and cardiometabolic disease [5, 6].

Chemerin is a recently discovered adipokine that may regulate adipocyte differentiation and metabolism [79]. It induces insulin resistance, exacerbates glucose intolerance, and decreases glucose uptake in skeletal muscle both in vitro and in vivo [1012]. In addition, elevated adipose tissue and serum chemerin levels have been observed in animal models of obesity and T2DM [7, 11].

In humans, elevated chemerin concentrations have been found in overweight and obese when compared to normal-weight individuals [1316] as well as in subjects with impaired glucose tolerance (IGT) [17] and T2DM [1315, 17, 18] when compared no normoglycemic subjects. Significant associations between serum chemerin levels and various components of the MetS have been observed in numerous studies [7, 1315, 19]. Therefore, chemerin has been recently suggested to mediate the link between obesity, inflammation, insulin resistance, and T2DM [2022].

Physical activity (PA) reduces adiposity, improves metabolic profile and cardiometabolic disease risk [23]. We recently demonstrated that a combined aerobic and resistance exercise program, in accordance with the current PA guidelines, may induced substantial beneficial changes in numerous cardiometabolic disease risk factors such as body mass index (BMI), waist circumference, fat mass, insulin resistance, blood pressure, and subclinical inflammation [24]. Data on the effect of exercise on chemerin levels is, however, sparse. Very recently, Venojärvi et al. [25] reported significant reduction in circulating chemerin as a result of both Nordic walking and resistance exercise intervention in men with impaired glucose regulation. Similarly, Chakaroun et al. [15] observed significant reduction in chemerin levels after 1 and 3 months of aerobic exercise training in individuals with (pre)diabetes. The authors, however, did not detect an exercise effect on chemerin levels in normoglycemic individuals, most likely because of the relatively short duration of the intervention. Furthermore, the effect of a program combining endurance and resistance exercises on circulating chemerin has not yet been investigated.

Therefore, in this study we aimed at: (1) investigating the relationship between chemerin and various cardiometabolic risk factors among non-diabetic individuals with a wide age and BMI range; (2) testing the hypothesis that a 6-month combined exercise program would decrease serum chemerin in middle-aged, overweight or obese, non-diabetic individuals.

Materials and methods

Study design and subjects

Combating the MetS through PA (COMPACT) study was conducted at the International Scientific Institute, National Sports Academy, Sofia, Bulgaria. Detailed description of study design was previously published [24]. Through postings at the University campus and announcements on the web page of the University, on the Internet and in the local newspapers 151 individuals, aged 18–60 years were admitted to initial interview.

Involvement in the study was not permitted if any the following exclusion criteria was met: medical history of T2DM; presence of a cardiovascular disorder or abnormality: arrhythmias, congenital heart defects, ischemic heart disease; myocardial infarction or stroke in the last 6 months; poorly controlled hypertension (≥160/110 mmHg); presence of neurological, thyroid, hepatic, renal or musculoskeletal disease; presence of acute or chronic inflammatory disease; adherence to a weight-loss diet or body weight change of more than ±3 kg in the last 3 months; use of medications or nutritional supplements that may affect energy expenditure and body weight, lipid and carbohydrate metabolism; drug or alcohol abuse. Based on these criteria, 107 subjects out of the initially interviewed 141 were included in the baseline examination. Of them nine individuals turned out to have a so far undiagnosed T2DM with fasting blood glucose ≥7.0 mmol/l and were excluded from further participation.

The remaining subjects were eligible for participation and were included in the cross-sectional analysis. Thus, the relationship of chemerin and various cardiometabolic risk factors was analyzed in 98 (n = 98) individuals without T2DM and with a wide range of age (18–60 years) and BMI (19.2–54.4 kg/m2).

In order to test the hypothesis that a 6-month combined exercise program would decrease serum chemerin in middle-aged, overweight or obese, non-diabetic individuals an experimental, controlled study was performed. Eligible for participation were subjects from the cross-sectional study who were 40–60 years old, had BMI ≥25 kg/m2 and were physically inactive at the time of enrollment (lack of engagement in structured PA of moderate to high intensity for more than 30 min two times per week). Eighty five (n = 85) subjects fulfilled these criteria and were enrolled in two study groups matched for age, sex, and BMI—the combined endurance and resistance exercise program group (CEP, n = 55) or the control group (C, n = 30). Four subjects dropped out from CEP group and two subjects from group C were lost to follow-up. Thus, 51 subjects from CEP group and 28 from the C group completed the study and were included in the final analysis. In a secondary analysis, we included only participants from the CEP group who completed more than 80 % of the targeted exercise sessions (participated in 52 out of 66 training sessions) and compared the effect of exercise training on chemerin levels in 28 subjects from CEP to 28 subjects in C group.

The study conforms to the principles outlined in the Declaration of Helsinki. All subjects were provided with a detailed explanation of the study aims and gave their written informed consent prior to the baseline medical examination.

Exercise program

All participants were provided healthy lifestyle counseling emphasizing on the health benefits of regular PA and balanced nutrition at a specially designed lecture at baseline. In addition, every subject had a scheduled 1 h personal meeting with a dietician at which he/she was given individually tailored advice and was provided with a standardized low calorie diet (1,400 and 1,800 kcal/day, for women and men, respectively) consisting of 55 % of total energy intake from carbohydrates, 30 % from fat, and 15 % from protein. During the study, no additional dietary counseling was provided and neither the dietary adherence nor the change in dietary habits was evaluated.

After the initial healthy lifestyle counseling and the provision of the dietary advice the subjects allocated to the C group were not contacted for the next 6 months until the final study visit.

Subjects in the CEP group participated in an intensive 6-month physical exercise program in accordance with the current PA guidelines promoting engagement in at least 150 min of aerobic exercises in combination with two sessions of resistance exercises per week.

All training sessions incorporated endurance exercises (e.g., brisk walking, low-intensity running, aerobics, tae-bo, dancing), resistance exercise (e.g., free weight exercises, own body weight exercises, resistance bands exercises, abdominal crunches), and flexibility exercises (e.g., stretching of the major muscle groups).

Since subjects were physically inactive prior to enrollment, they participated in a 1-week exercise course consisting of two lectures and three practical sessions during which they got acquainted with the basic principles of sports training and the execution of different exercises. Subsequently, during the initial 10 weeks of the study participants were encouraged to perform two exercise sessions per week with low to moderate intensity (aerobic exercises at 50–60 % of the maximum heart rate; resistance exercise for the major muscle groups in two sets of 8–14 repetitions) and duration of 30–35 min. Thus, for weeks 1–10 of the study the overall amount of exercise was 70 min/week corresponding to 350 metabolic equivalent-minutes per week (MET-min/week) [26]. This included 35 min/week (190 MET-min/week) of aerobic exercise and 35 min/week (160 MET-min/week) of resistance exercise. For 10 consecutive weeks, the number of sessions was raised to 3/week and their intensity (aerobic exercises at 60–75 % of maximum heart rate; resistance exercises in three sets of 8–14 repetitions) and duration (45–60 min) were also increased. Hence, for week 11–21 of the study the overall amount of PA was 180 min/week corresponding to 1,048 MET-min/week, which included 120 min/week (840 MET-min/week) of aerobic exercise and 60 min/week (208 MET-min/wee) of resistance exercise. During the last 4 weeks of the study the goal was participation in four training sessions per week (45–60 min). The total amount of exercise for these 4 weeks was 240 min/week corresponding to 1,256 MET-min/week, which included 120 min/week (840 MET-min/week) of aerobic exercise and 120 min/week (416 MET-min/week) of resistance exercise. The overall aim was participation in 66 exercise sessions for the whole study duration.

Measurements

All measurements were performed at baseline and after the 6-month exercise program between 8:00 a.m. and 10:00 a.m. after a 10 h overnight fast. Subjects were asked to refrain from strenuous exercise as well as from food excess of deprivation 3 days prior to both study visits.

Anthropometry and blood pressure assessment

Anthropometric measurements were performed according to standard techniques with subjects wearing light clothing. Body weight was measured to the nearest 0.05 kg on an electronic scale (CAS DB II, Sensortronic Scales, Manukau, New Zealand). Height was measured to nearest 0.5 cm without shoes and the subject standing with his/her back toward the wall. BMI was calculated as the body weight (kg) divided by height squared (m2). Waist circumference was determined with a plastic tape at the midpoint between the lower rib margin and the iliac crest to the nearest 0.5 cm. Blood pressure was examined in a sitting position after a rest of at least 5 min. Two consecutive measurements were performed within 3 min and the second one was taken into consideration.

Body composition analysis

Body composition analysis was performed using the bio impedance technique on a TANITA BC 418 MA analyser (TANITA Corporation, Tokyo, Japan).

Sub-maximal aerobic capacity assessment

Sub-maximal aerobic capacity was assessed on a treadmill (Quasar Med 4.0, HP Cosmos, Nussdorf-Traunstein, Germany) using the Modified Bruce protocol. The test was started at initial treadmill speed of 2.7 km/h at 0 % gradient for 3 min. Thereafter, speed and incline were gradually increased every 3 min according to the protocol [27]. The test was performed until volitional exhaustion or until the subject reached 85 % of his/her maximal hearth rate. Maximal aerobic capacity (VO2max) was then estimated using the equation of Foster [28].

Blood sample analyses

Venous blood samples were obtained by venipuncture. Plasma (EDTA) and serum were separated by centrifugation (4,000 rpm for 8 min at 4 °C). Aliquots of plasma and serum were immediately stored at −40 °C until further analyses. Fasting glucose, lipid profile, and leucocytes count were determined using fresh material.

Leucocytes count was determined photometrically on a Diagon D5 Cell analyser (Diagon Ltd., Budapest, Hungary). Fasting glucose was measured in serum by the glucose-hexokinase method on an Integra 400 Plus analyser (Roche Diagnostics GmbH, Mannheim, Germany). Fasting serum total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride concentrations were measured enzymatically on an Integra 400 Plus analyser (Roche Diagnostics GmbH). Serum low-density lipoprotein (LDL) cholesterol concentration was calculated using Friedewald equation. Fasting insulin concentration was measured in serum by ELISA (Mecordia, Uppsala, Sweden; inter-assay variability: 2.9 %; intra-assay variability: 3.4 %), which shows no cross-reactivity with proinsulin. Insulin resistance was calculated from fasting serum glucose and fasting serum insulin using the homeostasis model assessment for insulin resistance (HOMA-IR) [29]. The prevalence of the MetS among study participants was determined using the criteria of the International Diabetes Federation (IDF) [30]. Serum high sensitive C-reactive protein (hs-CRP) concentration was measured by ELISA (ap Dianv, Turnhout, Belgium; inter-assay variability: 6.1 %; intra-assay variability: 5.1 %). Serum leptin was measured by ELISA (Mediagnost, Reutlingen, Germany; sensitivity: 0.2 ng/ml; reference range of normal leptin values at a BMI of 25 kg/m2: males: 1.2–8.9 ng/ml, females: 8–24 ng/ml; inter- and intra-assay variability: <7 %). Serum chemerin was measured by ELISA (Biovendor, Heidelberg, Germany; inter-assay variability: 8.3 %; intra-assay variability: 5.1 %).

Statistical analyses

Data are presented as n, mean ± SD, or percentage (%) as, respectively, indicated. The distribution of variables was assessed by the Kolmogorov–Smirnov test for homogeneity of variance and non-normally distributed variables were logarithmically transformed to approximate normal distribution. Independent sample Student’s t test was used to compare characteristics between men and women in cross-sectional analyses, baseline characteristics of the subjects in the exercise and the control group as well as the magnitude of change in serum chemerin concentration and VO2max between the two groups after the 6-month exercise program. Within group changes in chemerin levels and VO2max were compared using paired sample Student’s t test. Correlation analyses were performed using Pearson or Spearman correlation coefficient as appropriate. Multiple linear regression analysis was conducted to check whether chemerin is a determinant of any of the investigated cardiometabolic risk factors. Differences were considered statistically significant at p < 0.05. All statistical analyses were executed using SPSS for Windows, version 17.0 (SPSS Inc., Chicago, IL).

Results

Demographic, anthropometric, and metabolic characteristics of the participants included in the cross-sectional analyses (n = 98) are presented in Table 1. Subjects were middle-aged (45.5 ± 0.9 years) and had mean BMI in the obesity range (32.1 ± 0.7 kg/m2). Women were older, had significantly lower waist circumference, fat-free mass, serum triglyceride concentration, and VO2max as well as higher fat mass, HDL cholesterol, and leptin concentration when compared to men. No other differences were observed between men and women. Serum chemerin concentration also did not differ between genders.
Table 1

Anthropometric and metabolic characteristics of the participants included in the cross-sectional analysis

Characteristic

Total

Men

Women

p value

N

98

38

60

 

Age (years)

45.5 ± 7.5

42.9 ± 7.8

47.1 ± 7.2

0.021

Body mass index (kg/m2)

32.1 ± 6.9

32.1 ± 7.1

32.2 ± 6.8

NS

Waist circumference (cm)

102.6 ± 17.7

108.4 ± 19.8

99 ± 16.3

0.01

Fat mass (%)

35.5 ± 8.7

27.3 ± 9.4

40.7 ± 7.8

<0.001

Fat-free mass (%)

64.5 ± 7.5

72.6 ± 7.5

59.3 ± 7.6

<0.001

Prevalence of MetS (%)

41.7

44.1

37.8

NS

Total cholesterol (mmol/l)

5.6 ± 1.2

5.9 ± 1.3

5.5 ± 1.1

NS

LDL cholesterol (mmol/l)

3.5 ± 1.0

3.7 ± 1.0

3.4 ± 0.9

NS

HDL cholesterol (mmol/l)

1.5 ± 0.4

1.3 ± 0.4

1.6 ± 0.4

0.004

Triglycerides (mmol/l)

1.5 ± 0.8

1.8 ± 1.0

1.3 ± 0.6

0.003

Fasting glucose (mmol/l)

5.5 ± 0.6

5.6 ± 0.6

5.4 ± 0.5

NS

Fasting insulin (pmol/l)

9.7 ± 6.9

11.6 ± 7.8

8.6 ± 6.0

0.05

HOMA-IR

2.4 ± 1.9

2.9 ± 2.3

2.1 ± 1.6

NS

Systolic blood pressure (mmHg)

120.0 ± 17.0

131.0 ± 15.7

128.0 ± 17.8

NS

Diastolic blood pressure (mmHg)

86.0 ± 9.6

87.0 ± 9.5

85.0 ± 9.6

NS

VO2max (ml/kg min)

32.5 ± 11.1

38.7 ± 7.3

28.0 ± 11.4

<0.001

hs-CRP (mg/l)

3.0 ± 2.7

2.6 ± 2.4

3.2 ± 2.9

NS

Leucocytes count (GPt/l)

7.8 ± 2.3

7.8 ± 2.7

7.8 ± 2.1

NS

Leptin (ng/ml)

26.4 ± 14.6

15.4 ± 11.4

32.9 ± 17.4

<0.001

Chemerin (ng/ml)

208.2 ± 38.1

208 ± 36.9

210.7 ± 38.9

NS

Data are presented as n, mean ± SD, or % as, respectively, indicated

NS not significant, MetS metabolic syndrome according to the criteria of the International Diabetes Federation

In univariate analyses, we observed a significant correlation of serum chemerin with age, BMI, waist circumference, percentage fat mass, total cholesterol, LDL cholesterol, triglycerides, fasting glucose, fasting insulin, HOMA-IR, systolic and diastolic blood pressure, leucocytes count, and leptin. Serum chemerin was also significantly but inversely correlated to percentage fat-free mass (Table 2). The association between circulating chemerin and total cholesterol, triglycerides, fasting insulin, HOMA-IR, systolic blood pressure, hs-CRP, leucocytes count, and leptin was slightly attenuated but remained significant after adjustment for age and BMI as potential confounders in partial correlation analysis (Table 2). In multiple linear regression analysis, serum chemerin concentration was determined to be a significant predictor of HOMA-IR (β = 0.265, 95 % CI 0.001–0.009, p = 0.039) independently of other risk factors for insulin resistance, namely age, gender, BMI, hsCRP, triglycerides, and family history of obesity and T2DM (data not shown).
Table 2

Correlations of serum chemerin with various cardiometabolic risk factors

 

Unadjusted

Age and BMI adjusted

r

p value

r

p value

Age

0.327

0.001

  

Sex

−0.086

NS

  

Body mass index

0.400

<0.001

  

Waist circumference

0.428

<0.001

0.163

NS

Fat mass

0.428

<0.001

0.062

NS

Fat-free mass

−0.406

<0.001

−0.126

NS

Total cholesterol

0.306

0.003

0.224

0.040

LDL cholesterol

0.221

0.030

0.084

NS

HDL cholesterol

−0.049

NS

-0.027

NS

Triglycerides

0.516

<0.001

0.438

<0.001

Fasting glucose

0.241

0.020

0.134

NS

Fasting insulin

0.510

<0.001

0.399

<0.001

HOMA-IR

0.509

<0.001

0.391

<0.001

Systolic blood pressure

0.381

0.002

0.219

0.04

Diastolic blood pressure

0.324

<0.001

0.157

NS

VO2max

−0.197

NS

−0.142

NS

hs-CRP

0.454

<0.001

0.236

0.030

Leucocytes count

0.368

<0.001

0.212

0.047

Leptin

0.450

<0.001

0.280

0.008

Data represent Pearson correlation coefficient values and p values

NS not significant

Baseline characteristics of the subjects in the CEP group (n = 51) and the control group (n = 28) are depicted in Table 3. No differences were observed in any of the examined parameters between the groups at baseline.
Table 3

Baseline characteristics of participants enrolled in the 6-month exercise program

Characteristic

CEP group

Control group

p value

n (men/women)

51 (18/33)

28 (9/19)

NS

Age (years)

47.8 ± 5.5

46.8 ± 6.1

NS

Body weight (kg)

94.2 ± 23.1

98.7 ± 16.1

NS

Body mass index (kg/m2)

33.5 ± 6.2

34.0 ± 5.1

NS

Prevalence of MetS (%)

43.1

53.6

NS

Waist circumference (cm)

106.2 ± 15.0

107.7 ± 12.4

NS

Fat mass (%)

38.7 ± 7.3

38.8 ± 7.2

NS

Fat-free mass (%)

61.2 ± 7.2

61.0 ± 7.2

NS

Total cholesterol (mmol/l)

5.8 ± 1.2

6.0 ± 0.9

NS

LDL cholesterol (mmol/l)

3.6 ± 1.0

3.7 ± 0.7

NS

HDL cholesterol (mmol/l)

1.5 ± 0.4

1.4 ± 0.4

NS

Triglycerides (mmol/l)

1.6 ± 0.8

1.6 ± 0.9

NS

Fasting glucose (mmol/l)

5.6 ± 0.6

5.4 ± 0.5

NS

Fasting insulin (pmol/l)

10.4 ± 6.1

11.0 ± 7.0

NS

HOMA-IR

2.8 ± 1.6

2.8 ± 2.4

NS

Systolic blood pressure (mmHg)

132.0 ± 16.6

132.0 ± 16.6

NS

Diastolic blood pressure (mmHg)

87.0 ± 9.2

86.0 ± 8.1

NS

VO2max (ml/kg min)

32.3 ± 12

33.4 ± 8.5

NS

hs-CRP (mg/l)

3.3 ± 2.6

3.3 ± 3.0

NS

Leucocytes count (GPt/l)

8.2 ± 2.4

8.2 ± 1.9

NS

Leptin (ng/ml)

27.8 ± 14.5

32.6 ± 19.5

NS

Chemerin (ng/ml)

212.9 ± 37.0

216.5 ± 38.6

NS

Data are presented as n, mean ± SD, or % as, respectively, indicated

CEP combined exercise group, MetS metabolic syndrome according to the criteria of the International Diabetes Federation, NS not significant

After the 6-month exercise program combining endurance and resistance exercises, maximal aerobic capacity (VO2max) increased significantly in CEP group (6.89 ± 4.72 ml/kg min, p < 0,001), and remained almost unchanged in C group (−0.32 ± 2.5 ml/kg min, p > 0.05). The change in VO2max observed in CEP was significantly greater compared to C group (p < 0.001; Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs12020-013-0003-2/MediaObjects/12020_2013_3_Fig1_HTML.gif
Fig. 1

Changes in VO2max after the 6-month exercise program. Black bar, combined exercise group (CEP), white bar, control (C) group. Data are presented as mean ± SD. *P < 0.05 vs. control group; †P < 0.05 within the CEP group after the 6-month exercise program

As a result of the 6-month combined exercise program chemerin decreased significantly from baseline within the CEP group (−13.8 ± 13.2 ng/ml, p = 0.002), whereas only a slight change was observed in C group (−7.4 ± 10.4 ng/ml, p > 0.05). No statistically significant between-group changes in chemerin levels were observed (Fig. 2a).
https://static-content.springer.com/image/art%3A10.1007%2Fs12020-013-0003-2/MediaObjects/12020_2013_3_Fig2_HTML.gif
Fig. 2

Changes in serum chemerin concentration after the 6-month exercise program in A) all subjects from CEP compared to C group and B) subjects from CEP who completed more than 80 % of the exercise sessions compared to C group. Black bar, Combined exercise group (CEP), white bar, control (C) group. Data are presented as mean ± SD. *P < 0.05 vs. control group; †P < 0.05 within the CEP group after the 6-month exercise program

In addition, in a secondary analysis we evaluated the effect of the 6-month combined exercise program on chemerin levels only among individuals who complete more than 80 % of the targeted exercise sessions. Among these subjects from CEP group even greater decrease in chemerin level was observed (−22.3 ± 13.8 ng/ml, p < 0.001). The magnitude of change in chemerin among the more compliant participant from CEP group was significantly higher in comparison to C group (−22.3 ± 13.8 vs. −7.4 ± 10.4 ng/ml, p = 0.028, respectively, 0.001; Fig. 2b).

In univariate analysis, the change (Δ) in chemerin was significantly correlated with Δwaist circumference, Δ% fat mass, Δ% fat-free mass, ΔLDL cholesterol, and ΔHOMA-IR. However, no association between Δchemerin, ΔBMI, and Δtriglycerides was observed. Adjustment for Δwaist abolished all of the significant associations except the association between Δchemerin and ΔHOMA-IR (Table 4).
Table 4

Correlations of Δchemerin with changes in other cardiometabolic risk factors at the end of the 6-month physical activity intervention

 

Unadjusted

ΔWaist-adjusted

r

p value

r

p value

Age

0.106

NS

−0.063

NS

Sex

0.031

NS

0.082

NS

ΔBody mass index

0.190

NS

−0.143

NS

ΔWaist circumference

0.348

0.012

ΔFat mass

0.400

0.004

0.260

NS

ΔFat-free mass

−0.374

0.007

−0.211

NS

ΔLDL cholesterol

0.280

0.004

0.211

NS

ΔTriglycerides

0.230

NS

0.216

NS

ΔHOMA-IR

0.382

0.006

0.332

0.018

Data represent Pearson correlation coefficient values and p values

NS not significant

Discussion

The novel finding of this study was that a 6-month exercise program combining endurance and resistance exercises led to a statistically significant reduction in serum chemerin concentration in middle-aged, overweight or obese, non-diabetic individuals. In addition, in cross-sectional analysis we observed a significant, BMI-independent association between serum chemerin and various cardiometabolic risk factors. Thus, our results confirm previous findings suggesting that chemerin may mediate the link between obesity, inflammation, insulin resistance, T2DM, and cardiometabolic disease, and add to the current understanding of the effect of exercise training on serum chemerin levels.

In a cross-sectional analysis among individuals without history of T2DM we found a significant correlation between serum chemerin and various cardiometabolic risk factors. This is consistent with the results from previous studies [7, 14, 16, 19, 3135]. Thus, association of chemerin levels with age [7, 31, 32], BMI [7, 14, 16, 31, 3335], waist circumference [7, 33, 35], fat mass [7, 14, 16, 19], blood lipids [7, 14, 3134], fasting glucose [7, 14], fasting insulin [7, 14, 33], insulin resistance [7, 14, 33], blood pressure [7, 3234], and biomarkers of inflammation [7, 31, 34, 35] has been previously reported. In accordance with some [19, 33, 34], but not all [7, 14, 31, 32] of the available reports we did not observe gender difference in serum chemerin levels. In contrast to the majority of the observations [7, 14, 19, 31, 32, 35], however, we did not find a significant association between chemerin and HDL cholesterol.

In this study, the relationship between chemerin and most of the examined cardiometabolic risk factors remained significant after adjustment for age and BMI. This corroborates the results of other investigators reporting BMI-independent association of chemerin with various components of the MetS [7, 14, 3136] and with the MetS per se [19, 32, 34]. Thus, our findings are in accordance with the observed by Bozaoglu et al. [14], Shin et al. [33], and Chu et al. [19] significant association of chemerin with lipid profile even after adjustment for age and BMI as potential confounding factors. We also found that the association of chemerin with HOMA-IR and fasting insulin was not abolished after BMI-adjustment, which is consistent with another report [14] and supports previous data showing that insulin sensitive individuals exhibit lower chemerin concentration compared to their age- and BMI-matched insulin resistant counterparts [20]. It is also in line with the observation that insulin administration may increase chemerin levels in healthy subjects [37].

Recently, Ouwens et al. [22] investigated the association between chemerin levels, the hyperinsulinemic-euglycemic clamp-derived M-value for insulin sensitive and surrogate indices for insulin sensitivity. The authors found that Pearson correlation coefficients between the M-value for insulin sensitivity and fasting levels of chemerin performed better than conventional surrogate measures of insulin sensitivity and that only the relation between M-value and chemerin remained significant when adjusting for BMI and fasting insulin and concluded that serum chemerin may be a useful biomarker of insulin sensitivity [22]. In accordance with these findings in this study a multiple linear regression analysis revealed chemerin was an independent determinant of HOMA-IR among non-diabetic individuals event after controlling for well-established risk factors for insulin resistance.

In addition, we observed a BMI-independent correlation between chemerin and systolic blood pressure, which corroborates previous findings [7, 3133] and supports the proposed by Bozaoglu et al. [7] involvement of this cytokine in the regulation of blood pressure. Furthermore, in this study the association of chemerin with hs-CRP and leucocytes count, was not abolished after adjustment for BMI. This finding is in accordance with the recently observed by Catalan et al. [35] BMI-independent correlation between chemerin and another well-established inflammatory biomarker—TNF-α. It also supports the notion that chemerin may be involved in the modulation of inflammatory response in adipose tissue [38] or that it is at least a marker of low-grade systemic inflammation [13, 31].

We demonstrate here for the first time that a 6-month exercise program combining endurance and resistance type exercises may significantly decrease chemerin concentration in overweight or obese non-diabetic individuals. Very recently, Venojärvi et al. [25] reported statistically significant reduction in serum chemerin as a result of both 3-month Nordic walking and resistance exercise program among overweight or obese men with impaired glucose regulation. Similarly, Chakaroun et al. [15] found that circulating chemerin is reduced after, respectively, 1 and 3 months of endurance training in individuals with T2DM and IGT, but not in individuals with normal glucose tolerance (NGT). Subjects with NGT in the latter study had lower BMI and chemerin levels at baseline compared to the participants in our study, which may to a certain extent explain the discrepancy [15]. However, it is also very likely that the twice-shorter duration of the exercise training was not sufficient to induce changes in serum chemerin concentration in individuals with NGT.

In a previous paper, we reported that a combined 6-month endurance and resistance exercise program may induce substantial beneficial changes in body mass, body composition, waist circumference, insulin resistance (estimated by HOMA-IR) and other well-established cardiometabolic disease risk factors in the same population of middle-aged, overweight or obese, previously inactive, non-diabetic individuals [24]. In this study, the decrease in chemerin levels was significantly associated with changes in insulin resistance (HOMA-IR) independently of changes in waist circumference which is in accordance with the findings of Chakaroun et al. [15] and corroborates previous observation demonstrating body weight- and BMI-independent associations between chemerin and insulin resistance [20, 22, 37].

Besides physical exercise, bariatric surgery and dietary interventions are known to induce substantial weight loss and to improve metabolic profile and insulin sensitivity. In accordance with our results, Chakaroun et al. [15] reported significant reduction in chemerin levels as a result of a 6-month hypocaloric dietary intervention in obese subjects. Similarly, Blüher and collaborates [39] found a significant decrease in serum chemerin in response to a low-fat restricted-calorie diet, Mediterranean restricted-calorie diet, and low-carbohydrate non-restricted-calorie diet during the 6-month weight loss phase of the Dietary Intervention Randomized Controlled Trial (DIRECT). In addition, Chakaroun et al. [15] and Sell et al. [40] observed a substantial decrease in serum chemerin at, respectively, 3 [40] and 12 months [15, 40] following weight-loss surgery in obese patients. Furthermore, decreased chemerin levels were found even 18 months after bariatric surgery [41]. In contrast to these findings, however, no change in chemerin was observed 13 months after a Roux-en-Y gastric bypass in obese women [35].

Poor cardiorespiratory fitness is an independent predictor of MetS, T2DM, and CVD, while high aerobic capacity is associated with lower cardiometabolic disease prevalence [4244]. Substantial improvements in maximal aerobic capacity were previously observed among middle-aged, overweight, or obese individuals after, respectively, 12 weeks [45], 16 weeks [46], and 8 months [47, 48] of combined exercise. Accordingly, we found a significant increase in VO2max after 6 months of training.

In conclusion, among non-diabetic individuals serum chemerin was associated with various cardiometabolic risk factors independently of BMI, which confirms previous reports suggesting that chemerin may mediate the link between obesity, inflammation, insulin resistance, T2DM, and CVD. In addition, a 6-month exercise program combining endurance and resistance exercises significantly reduced serum chemerin in overweight or obese, middle-aged, non-diabetic individuals and these changes were associated with improved insulin resistance independently of changes in waist circumference.

Acknowledgments

The authors would like to thank all study participants and Daniela Kox, Manuela Prellberg, and Susan Berthold for their technical assistance.

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical standards

The authors of the current manuscript declare that all the experiments comply with the current laws of the country in which they were performed (Bulgaria). The study conforms to the principles outlined in the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of the National Sports Academy, Sofia, Bulgaria and all subjects gave their written informed consent prior to participation.

Copyright information

© Springer Science+Business Media New York 2013