Lack of association between body mass index and plasma ... - Nature

8 downloads 0 Views 390KB Size Report
Mar 1, 2011 - risk of obesity, and this is the long-term aim of our analysis. Adiponectin and ... 8 January 2011; published online 1 March 2011 ..... (Figure 4b).
International Journal of Obesity (2011) 35, 1487–1494 & 2011 Macmillan Publishers Limited All rights reserved 0307-0565/11 www.nature.com/ijo

ORIGINAL ARTICLE Lack of association between body mass index and plasma adiponectin levels in healthy adults S-M Kuo1,2 and MM Halpern1 1 Department of Exercise and Nutrition Sciences, University at Buffalo, Buffalo, NY, USA and 2Department of Biochemistry, University at Buffalo, Buffalo, NY, USA

Objectives: To test the hypothesis that obesity (increase in fat mass) independently affects the level of adipokines: adiponectin, tumor necrosis factor-a (TNFa) and interleukin (IL)-6. Methods: Publications in the past decade reporting adult plasma adiponectin, leptin, TNFa and/or IL-6 levels were compiled. Mean gender-specific values were extracted from studies that included medical screening to confirm physical health (43 groups, total 4852 subjects). Correlation analysis was conducted between adipokine levels and body mass index (BMI), a widely used estimate of adiposity. Results: For healthy lean to obese adults of both genders, no significant correlation between plasma adiponectin and BMI was detected. There was also no gender difference in plasma adiponectin level. In contrast, leptin levels showed a positive correlation with BMI in both genders, and women had significantly higher levels of plasma leptin consistent with a higher percentage of body fat. The proinflammatory cytokine TNFa failed to show correlation with BMI. Although IL-6 showed a positive correlation with BMI in women, the obesity-related increase was very limited. Conclusions: Data analysis based on studies performed on healthy adults did not support the hypothesis that obesity independently affects the plasma level of adiponectin and TNFa. Reported obesity-related changes in plasma adipokine levels may be a consequence of obesity-related metabolic disorders. Future studies are especially needed to understand the homeostasis of adiponectin. International Journal of Obesity (2011) 35, 1487 – 1494; doi:10.1038/ijo.2011.20; published online 1 March 2011 Keywords: adiponectin; TNFa; IL-6; leptin; gender; meta-analysis

Introduction Obesity is a state of increased adipocyte size and subsequently an increase in total fat mass.1 With the industrialization, obesity is becoming a large health burden because of its association with the risk for disorders such as dyslipidemia, insulin resistance, hypertension, and eventually atherosclerosis and other cardiovascular diseases.2–7 Although the risk is well established, the mechanisms leading to the risk are not clear. A change in the plasma adipokine profile in obese individuals, such as the levels of adiponectin and the proinflammatory cytokines, tumor necrosis factor-a (TNFa) and interleukin (IL)-6, was proposed to be a contributing factor. However, in some cases, individuals were classified as

Correspondence: Dr S-M Kuo, Department of Exercise and Nutrition Sciences, University at Buffalo, 15 Farber Hall, Buffalo, NY 14214, USA. E-mail: [email protected] Received 25 May 2010; revised 15 November 2011; accepted 8 January 2011; published online 1 March 2011

obese, but showed no disorders.8 Transgenic fatless mice, upon excessive weight gain, also developed type 2 diabetes.9 Examining the homeostasis of adipokines in healthy obese individuals thus is one step toward understanding the health risk of obesity, and this is the long-term aim of our analysis. Adiponectin and leptin are cytokines mainly produced from the white adipose tissue as transgenic fatless mice have little serum leptin and adiponectin.9 Leptin relayed fat storage size to the central nervous system to regulate food intake and energy storage.10 Rodents deficient in leptin or leptin receptor developed severe obesity and obesity-associated pathological conditions due to a disrupted communication between peripheral tissue and central nervous system.11,12 On the basis of the source and biological function of leptin, it was expected and indeed well established that the circulating level of leptin increases proportional to the degree of obesity.10 Circulating level of adiponectin is 100- to 1000-fold than that of leptin.10 In contrast to leptin knockout mice, adiponectin knockout mice fed regular chow were quite

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1488 normal.13,14 Under high-fat diet, knockout mice exhibited metabolic disturbance including elevated plasma free fatty acid levels and insulin resistance. However, the metabolic disturbance was independent of body weight as the body weight of adiponectin knockout mice was similar to that of the wild-type mice fed the same diet (regular or high-fat diet).13,14 These observations do not support a role of adiponectin in body weight regulation, which is different from leptin. Instead, adiponectin seems to be important for energy metabolism, especially the metabolism of free fatty acids. Epidemiological studies reported an association of lower plasma adiponectin levels with metabolic syndrome, type 2 diabetes15,16 and cardiovascular diseases.17 As the risk for these disorders increases with obesity, several types of studies tested the hypothesis that obesity depresses the plasma adiponectin level, which in turn increases the risk for diabetes and cardiovascular diseases. Weight reduction through gastric surgery18 or the removal of visceral fat19 led to the amelioration of insulin resistance and an increase in plasma adiponectin. However, plasma levels of adiponectin were shown to be more closely related to insulin resistance20,21 or type 2 diabetes22 than with adiposity. In fact, for healthy human subjects and animal, moderate weight loss did not necessarily translate to changes in the plasma adiponectin level.23,24 For subjects with metabolic abnormalities, weight loss larger than 10% was needed to change plasma adiponectin levels.25 Also, patients with defective growth hormone receptor showed elevated plasma adiponectin despite marked obesity.26 Limitations such as interventions applied, small sample size and disease conditions (or unclear health status) were often observed in other human studies and thus compromised the ability to conclude that obesity directly suppresses the plasma adiponectin level. To examine whether obesity independently depresses the plasma level of adiponectin, we performed pooled data analysis using published studies on clinically evaluated nonintervened healthy subjects. Body mass index (BMI) was used to quantify the degree of obesity. Partly because of the easiness in obtaining the information, BMI values were available in almost all literatures that examined adiposity. BMI was also shown to be as effective as other anthropometric variables in estimating adiposity-related outcome based on the National Health and Nutrition Examination Survey III data analysis.27 We limited the analysis to those studies providing gender-specific data because women are known to have higher percentage body fat than men at the same BMI.28–30 To validate the quality of our analyses, plasma leptin level, which was known to increase with obesity and found to be higher in women,31 was also analyzed here as a positive control. The same BMI correlation analyses were also applied to two proinflammatory cytokines, IL-6 and TNFa. They are mainly from macrophages, but also secreted by adipocytes.32 Obesity is considered a low-grade inflammatory state33 and proinflammatory cytokines were thought to be involved in International Journal of Obesity

the development of obesity-related insulin resistance and cardiovascular diseases. However, not all studies found a correlation between plasma proinflammatory cytokine levels and obesity.34 Similar to adiponectin, the secretion of proinflammatory cytokines was known to be affected by diabetes35,36 and cardiovascular diseases.37–39 In addition, their levels rose on infection, inflammation or acute stress.40 To determine the independent effect of obesity on their levels, our analysis was again restricted to data from studies that enrolled clinically evaluated non-intervened healthy subjects.

Materials and methods Data collection Data used for the analysis were collected from published human studies in PubMed database. Publications within the past decade (2000 to May, 2009) that included serum/plasma adiponectin, leptin, TNFa and/or IL-6 measurements were first compiled. None of the studies measured all four cytokines. Nevertheless, all studies reported BMI, except one that provided mean weight/height values, thus permitting the calculation of mean BMI. To avoid the confounding effect of diseases and metabolic abnormalities, studies that did not include health screening and those that included subjects with diabetes, cardiovascular diseases or other illness were excluded from the analysis. For studies involving interventions such as caloric reduction and/or exercise, only the pre-intervention basal values of the subjects were used for the analysis to avoid the confounding effect of intervention. Thus, each data used for the analysis represented an observation made on a distinct group of individuals. Furthermore, a few studies with more than 10-fold higher TNFa or IL-6 values than the mean of other studies were excluded. Abnormally high baseline values of these inflammatory cytokines may be a result of minor infection or underlying undiagnosed pathological conditions.41–43 Also excluded are a few studies reporting values of leptin or adiponectin that are at least 10-fold different, higher or lower, from the mean of other publications. Of the remaining studies, because of the known continuous body composition changes from birth to adulthood (18 years old),44 only studies on subjects older than 18 years of age were included in the analysis. Furthermore, studies reporting mean values combining observations from both genders were not used because of the known gender difference in percentage body fat,28–30 fat distribution29 and plasma leptin level.31 Overall, the three exclusion criteria, concerns on subject health, young age and a lack of gender focus, led to the majority of the data exclusion. Male data were compiled from 13 studies (17 groups of individuals), with group sample size ranging from 10 to 466 individuals and a mean group sample size of 142.29,34,45–55 Female data were compiled from 15 studies (26 groups of individuals), with study sample size ranging from 8 to 662 individuals and a mean sample size of 94.29,45,46,48,50,53–62

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1489 Table 1

Sample size, age and BMI profiles of studies used for the analysis

Sample size

Age

BMI

Range (years)

Mean±s.d. (years)

Mean±s.d.

Male studies 466 32 208 452 223 59 21 54 12 10 282 163 163 198 16 16 41

41–52 18–25 19–72 18–81 18–81 450 F F F F 30–61 30–59 30–59 30–65 18–35 18–35 30–49

47±3 21±2 42±15 40±14 42±14 59±8 19±1 56±1 35±4 36±3 46±9 45±8 45±8 44±9 30±2 31±2 39±5

27.9±1.2 23.4±2.4 26.6±4.3 27.7±1.6 28.1±1.5 25.6±3.6 31.1±4.2 23.3±0.4 28.9±1.3 28.8±1.8 23.2±2.7 22.0±2.1 26.6±2.8 22.9±2.2 24.5±0.6 34.1±1.1 26.3a

Female studies 392 62 145 67 662 299 28 105 108 8 8 15 14 221 21 81 24 28 15 14 38 15 16 9 21 20

45–57 18–25 18–61 F 18–81 18–81 F 20–41 20–41 18–23 18–23 30–70 30–70 30–65 F F F F 18–35 18–35 30–47 50–70 50–70 50–70 F F

51±3 21±2 37±11 49±12 42±13 40±13 58±2 34±6 34±6 18±1 18±2 54±9 53±10 42±10 41±10 41±11 33±11 29±9 28±2 28±2 39±5 56±1 59±1 57±2 43±11 42±9

26.4±2.0 22.9±2.9 26.0±5.2 32.4±4.5 27.5±2.1 28.7±2.7 24.0±1.0 28.2±1.4 28.3±1.3 29.5±2.7 21.9±3.2 37.3±5.5 22.9±1.9 20.6±2.4 44.7±4.0 34.4±2.8 28.1±1.6 21.8±2.0 23.0±0.8 39.0±1.8 26.6a 33.6±1.2 31.1±1.0 31.4±1.2 46.0±8.6 45.2±5.2

Refs.

45 46 29

were reported as means and data ranges instead of means and standard deviations.45,48,55 For the purpose of data plotting, we calculated standard deviations for those studies by dividing the difference between the reported maximum and minimum values by 4. This estimation did not impact on the correlation analysis. Gender difference was determined by two-tailed Student’s t-test and significant difference was defined as Po0.05.

48 48 47

Results

49 50 51 51 52 34 34 53 54 54 55

45 46 29 56 48 48 50 57 57 58 58 59 59 53 60 60 60 60 54

Profile of studies included in the analysis Table 1 summarizes the sample size, age range, mean age and mean BMI of the studies used in the analysis. The mean age of male groups ranged from 19 to 59 years, which was similar to that of female groups that ranged from 18 to 59 years. The mean BMI ranged from 22.0 to 34.1 kg m 2 for male groups and 20.6 to 46.0 kg m 2 for female groups. By definition, BMI between 18.5 and 24.9 kg m 2 is considered normal, between 25.0 and 29.9 kg m 2 overweight, BMI above 30 kg m 2 as obese and BMI greater than 40 kg m 2 as morbid obesity. Thus, our analysis covered various measurements made on adults of both genders ranging from normal weight to obese. Although all male groups had mean BMI of less than 35 kg m 2, three of 26 female groups had mean BMI between 35 and 40 kg m 2 and two of 26 had mean BMI greater than 40 kg m 2 in the range of morbid obesity. No significant gender difference in BMI was detected by Student’s t-test. In the following data analysis, mean BMI was used as the independent variable to represent the degree of obesity of the subject group. The linear relationship between each dependent variable, mean plasma adiponectin, leptin, IL-6 or TNFa, with mean BMI, was analyzed to see if BMI can predict the plasma level of these adipokines. In addition, we used Student’s t-test to examine whether there was gender difference in mean plasma levels of adiponectin, leptin, IL-6 or TNFa.

54 55 61 61 61 62 62

Abbreviation: BMI, body mass index; F, data not available. aCalculated from the mean weight and height of the group.

Among these studies, eight studies included subjects of both genders and the measurements from men and women were reported separately. The sample size as well as the age and BMI profiles of these studies are summarized in Table 1.

Data analysis Mean values of adipokines and BMI in each study were used for the linear regression analysis and the computation of Pearson’s correlation coefficient. Significant correlation was defined as Po0.05. In three studies, experimental results

Analyses of adiponectin and leptin Linear regression of plasma adiponectin against BMI was performed separately for men (Figure 1a) and women (Figure 1b). The regression analysis of male adiponectin to BMI was based on data from 12 groups34,45,47–50,52–54 and Pearson’s correlation was 0.26 (P40.05, not significant). Thus, there was no correlation between mean plasma adiponectin and BMI in men (Figure 1a). The regression analysis of female adiponectin to BMI was based on data from 17 groups45,48,50,53,54,56–58,61,62 and Pearson’s correlation was 0.52 (Po0.05). However, the borderline significant negative correlation of plasma adiponectin to BMI in women disappeared if the two female groups with morbid obesity (mean BMI of 45.2 and 46 kg m 2) were excluded (Pearson’s correlation 0.38, P40.05). Within the BMI range of normal to obese, there was no correlation between plasma adiponectin and BMI. Adiponectin is known to form different structural complexes ranging from homotrimer (low-moleInternational Journal of Obesity

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1490

Figure 1 Relationship between baseline plasma adiponectin and BMI in

Figure 2 Relationship between baseline plasma leptin and BMI in healthy

healthy adult men (a) and women (b). Mean and standard deviation of plasma adiponectin and the corresponding mean BMI were compiled from various male34,45,47–50,52–54 and female45,48,50,53,54,56–58,61,62 studies included in Table 1. Some error bars are too small to be visible.

adult men (a) and women (b). Mean and standard deviation of plasma leptin and the corresponding mean BMI were compiled from various male45–47,49–51,53,54 and female45,46,50,53,54,56,58,61,62 studies included in Table 1. Some error bars are too small to be visible. Po0.01 for the correlation between plasma leptin and BMI in men and Po0.05 for the leptin–BMI correlation in women. Po0.01 for a gender difference in leptin.

cular weight) to high-molecular weight 16–18mers.63,64 Because adiponectin levels were determined by enzymelinked immunosorbent assay or radioimmunoassay, which by nature could not distinguish different structural forms, the BMI–adiponectin regression analysis was based on the total adiponectin. Specialized enzyme-linked immunosorbent assay that only measured high-molecular weight adiponectin was used in one male study.52 Removing the high-molecular weight data point from the regression analysis did not affect the conclusion of no BMI–adiponectin correlation. The results of plasma leptin to BMI regression analysis are shown in Figure 2. The regression analysis of male leptin to BMI was based on data from 10 groups (Figure 2a)45–47,49–51,53,54 and Pearson’s correlation was 0.79 (Po0.05). The regression analysis of female leptin to BMI was based on data from 14 groups (Figure 2b)45,46,50,53,54,56,58,61,62 and Pearson’s correlation was 0.58 (Po0.05). The leptin–BMI correlation was consistent with the published result. Unlike adiponectin, the significant leptinBMI correlation was not affected by the exclusion of subjects with morbid obesity, BMI over 40 kg m 2 (Pearson’s correlation 0.77, Po0.05). The plasma level of leptin was shown to be significantly higher in women compared with men.31 This gender difference in plasma leptin level was also significant in our data analysis (mean±s.d., 7.75±7.13 vs 22.5±12.2 ng ml 1, M vs F). In contrast, the plasma level of adiponectin was not significantly different between men and women in our data analysis (mean±s.d., 7.34±2.74 vs 9.31±4.81 mg ml 1, M vs F). On the basis of the positive correlation between plasma leptin and BMI (Figure 2) and no correlation between adiponectin and BMI (Figure 1), we expected no correlation between plasma adiponectin and leptin in healthy individuals. Some studies we compiled reported both plasma leptin and adiponectin measurements, thus allowing the correlation analysis between adiponectin and leptin. The regression analysis of male adiponectin to leptin was performed on data from seven groups45,47,49,50,53,54 and Pearson’s correlation was 0.53 (P40.05, not significant). International Journal of Obesity

Figure 3 Relationship between baseline plasma TNFa and BMI in healthy adult men (a) and women (b). Mean and standard deviation of plasma TNFa and the corresponding mean BMI were compiled from various male29,55 and female29,55,57–60 studies included in Table 1. Some error bars are too small to be visible.

The regression analysis of female adiponectin to leptin was based on data from 13 groups45,50,53,54,56,58,61,62 and Pearson’s correlation was 0.37 (P40.05, not significant). The analysis confirmed the expected lack of correlation between adiponectin and leptin in healthy individuals.

Analyses of TNFa and IL-6 The plasma TNFa and IL-6 data that we compiled are shown in Figures 3a and 4a (men) and Figures 3b and 4b (women). After applying our data exclusion criteria, we found very limited information available on the plasma TNFa level of healthy adult men (Figure 3a),29,55 so regression analysis was not possible. The regression analysis of female TNFa to BMI was based on data from 12 groups (Figure 3b)29,55,57–60 and Pearson’s correlation was 0.51 (P40.05, not significant). There was also limited information available (six groups) on the plasma IL-6 level of healthy adult men (Figure 4a).29,34,45,46 The Pearson’s correlation was 0.62 for the male IL-6 to BMI correlation analysis (P40.05, not significant). The regression analysis of female IL-6 to BMI was based on data from eight groups

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1491

Figure 4 Relationship between baseline plasma IL-6 and BMI in healthy adult men (a) and women (b). Mean and standard deviation of plasma IL-6 and the corresponding mean BMI were compiled from various male29,34,45,46 and female29,45,46,55,57,59 studies included in Table 1. Some error bars are too small to be visible. Po0.01 for the correlation between plasma IL-6 and BMI in women.

(Figure 4b).29,45,46,55,57,59 The Pearson’s correlation was 0.85 (Po0.05), indicating a significant positive correlation between plasma IL-6 and BMI in healthy adult women. No gender differences were found in plasma levels of TNFa and IL-6. The plasma level of TNFa in healthy adult women was 3.55±2.72 pg ml 1 (mean±s.d.), which was similar to the mean plasma level of TNFa from two healthy adult male groups (2.90 pg ml 1). The plasma levels of IL-6 in healthy adult men and women were 1.20±0.21 and 1.51±0.48 pg ml 1, respectively (mean±s.d.).

Discussion We conducted pooled data analysis on healthy subjects to determine whether obesity, in the absence of metabolic disorders, independently affected the plasma level of adipokines. Although plasma leptin level, our positive control, showed a positive correlation with BMI (Figure 2), we did not find a correlation between BMI and plasma levels of adiponectin or TNFa in the same population of healthy normal to obese adults (Figures 1 and 3). Although a positive correlation was found between BMI and IL-6 in women (Figure 4b), the magnitude of changes was limited. Women are known to have higher body fat content than men and, as expected, the positive control leptin showed significantly higher levels in women than men (Figure 2). In contrast, the levels of adiponectin, TNFa and IL-6 did not exhibit gender differences, which was consistent with their lack of correlation with adiposity. Gender-specific pooled data analysis of adipokines in healthy adults has never been reported before. Our finding on the lack of correlation between BMI and adiponectin level in healthy adults strengthened a similar conclusion drawn from subjects with various degrees of insulin resistance.20,65 Although adiponectin is known to have different structural complexes, low-molecular weight, medium molecular weight and highmolecular weight,63,64 total adiponectin was reported in almost all studies so that the effect of BMI on the

distribution of adiponectin complexes cannot be determined. The distribution of three complexes is known to be affected by gender and by insulin resistance.64 A recent study on women with polycystic ovary syndrome found that the levels of three adiponectin complexes changed largely in parallel to each other.66 Our findings on adiponectin support the presence of tight regulation of this adipokine. Adiponectin resistance has been observed.67 In fact, high fat-fed rats developed skeletal muscle adiponectin resistance before the onset of insulin resistance.67 Also, transgenic adiponectin-overexpressing mice eventually had lower circulating adiponectin level than that of wild-type mice indicating the presence of feedback regulation. 68 It is known that changes in gene expression cannot always explain the changes in circulating adiponectin levels.69 Thus, a simple expansion of adipose tissue in healthy obese individuals should not be expected to affect the plasma level of adiponectin and the results of our analysis supported this (Figure 1). Instead, regulation at the rate of adiponectin secretion may happen under various obesity-related pathological conditions such as the metabolic syndrome,70 which then lead to a lower plasma adiponectin levels.15–17,71 Metabolic syndrome, because of its association with the infiltration of macrophages, is considered a state of tissue inflammation.71 Conditions found in the metabolic syndrome such as high plasma glucose,72 high dietary fat,73 changes in very-low-density lipoprotein metabolism74 and inflammatory cytokines70,75 are all known to inhibit adiponectin secretion in vitro. Our analysis cannot provide direct information on the biochemicals involved in the physiological regulation of adiponectin secretion. However, the results of our analysis were consistent with a role of free fatty acids, specifically an elevation of circulating free fatty acids led to the decrease in the circulating level of adiponectin. Although plasma free fatty acid levels were not always measured in human studies, elevated levels of free fatty acid were found in many obesityrelated pathological conditions including type 2 diabetes76,77 and metabolic syndrome.78 Physiological changes due to an elevation in plasma free fatty acids are known and lipotoxicity could be a link between obesity and type 2 diabetes.79,80 An elevation of plasma free fatty acids is known in experimental low-dose endotoxemia, a human non-obese model to study inflammatory response.81 It is possible that some obese individuals remained healthy because of a higher capacity to hold storage fat due to genetic and/or environmental factors.80 Thus, their fatty acid levels did not increase and adiponectin levels did not decrease. Female subjects in our analysis were able to stay healthy at a higher BMI than male subjects (Figures 1–4). This may be due to their higher capacity for adipose tissue expansion, likely a gender-specific evolutionary advantage for reproductive success. It is interesting that some other pathological conditions were associated with an elevation of circulating adiponectin.69 At least one of the conditions, anorexia nervosa, is known to lead to abnormal lipid metabolism.82 International Journal of Obesity

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1492 Future human studies correlating the plasma level of free fatty acids (and perhaps even the type of free fatty acids83,84) and that of adiponectin are needed to further examine the potential free fatty acid–adiponectin link. Results from correlation studies performed on healthy human studies suggest that such a link exist.85,86 Although acute druginduced reduction of plasma free fatty acid in healthy human subjects was associated with a reduction of plasma adiponectin,87 niacin treatment that decreased plasma free fatty acids led to a rapid increase in the plasma adiponectin concentration in healthy rats and mice likely through an increased secretion.88 Studies on animal models provided additional information on the link between adiponectin and free fatty acids. Adiponectin appeared to be important for the metabolism of fatty acids as shown in mouse models.13,14,89 The absence of adiponectin coincided with a higher plasma free fatty acid level.13Adiponectin treatment dose-dependently decreased plasma free fatty acid levels.89 Rats given high-fat diet, which is known to increase the plasma level of free fatty acids,13,79 developed skeletal muscle adiponectin resistance before insulin resistance67 and the resistance could not be explained by muscle inflammation.83 Our analysis revealed a significant correlation between BMI and plasma IL-6 (Figure 4b), but not between BMI and plasma TNFa (Figure 3b). This is consistent with the in vitro observation on adipocytes where larger adipocytes were shown to release more IL-6, but not more TNFa.90 However, the physiological significance of the positive correlation between BMI and IL-6 is not clear. IL-6 is estimated to increase from about 1 pg ml 1 at BMI of 22 mg kg 2 to about 2.5 pg ml 1 at BMI of 37 mg kg 2 (a state of severe obesity) (Figure 4b). This increase was very limited in amount compared with the IL-6 level of post-cardiac disease,91 post-coronary artery bypass surgery patients92 or experimental low-dose entotoxemia in healthy human subjects.81 It is not convincing that the slight IL-6 elevation can contribute to the development of obesity-related metabolic diseases. Several recent reviews also reached the same conclusion.93–95 In summary, our gender-specific analysis on healthy adults did not support an association between adiposity and plasma adiponectin level despite a significant positive correlation between adiposity and leptin, another white adipose tissue cytokine. A constant plasma adiponectin level in healthy normal to obese subjects suggested the presence of regulatory mechanisms. Because of the known biological importance of adiponectin and an observed decrease of adiponectin level in metabolic diseases,15–17,71 understanding the regulatory mechanisms for adiponectin would help the management of metabolic diseases.

Conflict of interest The authors declare no conflict of interest. International Journal of Obesity

Acknowledgements The statistical consultation provided by Dr Yow-Wu Wu of University at Buffalo, School of Nursing, is greatly appreciated.

References 1 Hirsch J, Batchelor B. Adipose tissue cellularity in human obesity. Clin Endocrinol Metab 1976; 5: 299–311. 2 Moayyedi P. The epidemiology of obesity and gastrointestinal and other diseases: an overview. Dig Dis Sci 2008; 53: 2293–2299. 3 Markoff B, Amsterdam A. Impact of obesity on hospitalized patients. Mt Sinai J Med 2008; 75: 454–459. 4 Ben-Sefer E, Ben-Natan M, Ehrenfeld M. Childhood obesity: current literature, policy and implications for practice. Int Nurs Rev 2009; 56: 166–173. 5 Mathieu P. Abdominal obesity and the metabolic syndrome: a surgeon’s perspective. Can J Cardiol 2008; (Suppl D): 19D–23D. 6 Ervin RB. Prevalence of metabolic syndrome among adults 20 years of age and over, by sex, age, race and ethnicity, and body mass index: United States, 2003–2006. Natl Health Stat Report 2009; 5: 1–7. 7 Vgontzas A, Papanicolaou D, Bixler E, Hopper K, Lotsikas A, Lin H et al. Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. J Clin Endocrinol Metab 2000; 85: 1151–1158. 8 Harrington M, Gibson S, Cottrell R. A review and meta-analysis of the effect of weight loss on all-cause mortality risk. Nutr Res Rev 2009; 22: 93–108. 9 Nunez N, Oh W, Rozenberg J, Perella C, Anver M, Barrett J et al. Accelerated tumor formation in a fatless mouse with type 2 diabetes and inflammation. Cancer Res 2006; 66: 5469–5476. 10 Dyck D. Adipokines as regulators of muscle metabolism and insulin sensitivity. Appl Physiol Nutr Metab 2009; 34: 396–402. 11 Pelleymounter M, Cullen M, Baker M, Hecht R, Winters D, Boone T et al. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 1995; 269: 540–543. 12 De Artinano A, Castro M. Experimental rat models to study the metabolic syndrome. Br J Nutr 2009; 102: 1246–1253. 13 Maeda N, Shimomura I, Kishida K, Nishizawa H, Matsuda M, Nagaretani H et al. Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 2002; 8: 731–737. 14 Kubota N, Terauchi Y, Yamauchi T, Kubota T, Moroi M, Matsui J et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 2002; 277: 25863–25866. 15 Yatagai T, Nagasaka S, Taniguchi A, Fukushima M, Nakamura T, Kuroe A et al. Hypoadiponectinemia is associated with visceral fat accumulation and insulin resistance in Japanese men with type 2 diabetes mellitus. Metabolism 2003; 52: 1274–1278. 16 Li S, Shin H, Ding E, van Dam R. Adiponectin levels and risk of type 2 diabetes: a systematic review and meta-analysis. JAMA 2009; 302: 179–188. 17 Kumada M, Kihara S, Sumitsuji S, Kawamoto T, Matsumoto S, Ouchi N et al. Association of hypoadiponectinemia with coronary artery disease in men. Arterioscler Thromb Vasc Biol 2003; 23: 85–89. 18 Yang W, Lee W, Funahashi T, Tanaka S, Matsuzawa Y, Chao C et al. Weight reduction increases plasma levels of an adipose-derived anti-inflammatory protein, adiponectin. J Clin Endocrinol Metab 2001; 86: 3815–3819. 19 Pitombo C, Araujo E, De Souza C, Pareja J, Geloneze B, Velloso L. Amelioration of diet-induced diabetes mellitus by removal of visceral fat. J Endocrinol 2006; 191: 699–706. 20 Weyer C, Funahashi T, Tanaka S, Hotta K, Matsuzawa Y, Pratley R et al. Hypoadiponectinemia in obesity and type 2 diabetes: close

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1493 21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

association with insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab 2001; 86: 1930–1935. Mojiminiyi OA, Abdella NA, Al Arouj M, Ben Nakhi A. Adiponectin, insulin resistance and clinical expression of the metabolic syndrome in patients with Type 2 diabetes. Int J Obes Relat Metab Disord (Lond) 2007; 31: 213–220. Nakashima R, Kamei N, Yamane K, Nakanishi S, Nakashima A, Kohno N. Decreased total and high molecular weight adiponectin are independent risk factors for the development of type 2 diabetes in Japanese-Americans. J Clin Endocrinol Metab 2006; 91: 3873–3877. Mazzali G, Di Francesco V, Zoico E, Fantin F, Zamboni G, Benati C et al. Interrelations between fat distribution, muscle lipid content, adipocytokines, and insulin resistance: effect of moderate weight loss in older women. Am J Clin Nutr 2006; 84: 1193–1199. Romero MM, Ferna´ndez-Lo´pez J, Esteve M, Alemany M. Different modulation by dietary restriction of adipokine expression in white adipose tissue sites in the rat. Cardiovasc Diabetol 2009; 8: 42. Madsen E, Rissanen A, Bruun J, Skogstrand K, Tonstad S, Hougaard D et al. Weight loss larger than 10% is needed for general improvement of levels of circulating adiponectin and markers of inflammation in obese subjects: a 3-year weight loss study. Eur J Endocrinol 2008; 158: 179–187. Kanety H, Hemi R, Ginsberg S, Pariente C, Yissachar E, Barhod E et al. Total and high molecular weight adiponectin are elevated in patients with Laron syndrome despite marked obesity. Eur J Endocrinol 2009; 161: 837–844. Flegal K, Graubard B. Estimates of excess deaths associated with body mass index and other anthropometric variables. Am J Clin Nutr 2009; 89: 1213–1219. Chumlea W, Guo S, Kuczmarski R, Flegal K, Johnson C, Heymsfield S et al. Body composition estimates from NHANES III bioelectrical impedance data. Int J Obes Relat Metab Disord 2002; 26: 1596–1609. Cartier A, Cote M, Lemieux I, Perusse L, Tremblay A, Bouchard C et al. Sex differences in inflammatory markers: what is the contribution of visceral adiposity? Am J Clin Nutr 2009; 89: 1307–1314. Li C, Ford E, Zhao G, Balluz L, Giles W. Estimates of body composition with dual-energy X-ray absorptiometry in adults. Am J Clin Nutr 2009; 90: 1457–1465. McConway M, Johnson D, Kelly A, Griffin D, Smith J, Wallace A. Differences in circulating concentrations of total, free and bound leptin relate to gender and body composition in adult humans. Ann Clin Biochem 2000; 37: 717–723. Wang B, Wood I, Trayhurn P. Dysregulation of the expression and secretion of inflammation-related adipokines by hypoxia in human adipocytes. Pflugers Arch 2007; 455: 479–492. Valle M, Martos R, Gascon F, Canete R, Zafra MA, Morales R. Lowgrade systemic inflammation, hypoadiponectinemia and a high concentration of leptin are present in very young obese children, and correlate with metabolic syndrome. Diabetes Metab 2005; 31: 55–62. Nishida M, Moriyama T, Sugita Y, Yamauchi-Takihara K. Abdominal obesity exhibits distinct effect on inflammatory and antiinflammatory proteins in apparently healthy Japanese men. Cardiovasc Diabetol 2007; 6: 27. Tsai G, Cui J, Syed H, Xia Z, Ozerdem U, McNeill J et al. Effect of N-acetylcysteine on the early expression of inflammatory markers in the retina and plasma of diabetic rats. Clin Exp Ophthalmol 2009; 37: 223–231. Bradshaw E, Raddassi K, Elyaman W, Orban T, Gottlieb P, Kent S et al. Monocytes from patients with type 1 diabetes spontaneously secrete proinflammatory cytokines inducing Th17 cells. J Immunol 2009; 183: 4432–4439. Picciotto S, Forastiere F, Pistelli R, Koenig W, Lanki T, Ljungman P et al. Determinants of plasma interleukin-6 levels among

38

39

40

41 42 43 44

45

46

47

48

49

50

51

52

53

54

55

survivors of myocardial infarction. Eur J Cardiovasc Prev Rehabil 2008; 15: 631–638. Licata G, Tuttolomondo A, Di Raimondo D, Corrao S, Di Sciacca R, Pinto A. Immuno-inflammatory activation in acute cardioembolic strokes in comparison with other subtypes of ischaemic stroke. Thromb Haemost 2009; 101: 929–937. Tuttolomondo A, Di Sciacca R, Di Raimondo D, Serio A, D’Aguanno G, La Placa S et al. Plasma levels of inflammatory and thrombotic/fibrinolytic markers in acute ischemic strokes: relationship with TOAST subtype, outcome and infarct site. J Neuroimmunol 2009; 215: 84–89. Gray SR, Robinson M, Nimmo MA. Response of plasma IL-6 and its soluble receptors during submaximal exercise to fatigue in sedentary middle-aged men. Cell Stress Chaperones 2008; 13: 247–251. Kishimoto T. Interleukin-6: from basic science to medicineF40 years in immunology. Annu Rev Immunol 2005; 23: 1–21. Lacour S, Gautier J, Pallardy M, Roberts R. Cytokines as potential biomarkers of liver toxicity. Cancer Biomark 2005; 1: 29–39. Tisdale M. Mechanisms of cancer cachexia. Physiol Rev 2009; 89: 381–410. Kelly T, Wilson KE, Heymsfield S. Dual energy X-ray absorptiometry body composition reference values from NHANES. PLoS One 2009; 4: e7038. Badellino K, Wolfe M, Reilly M, Rader D. Endothelial lipase is increased in vivo by inflammation in humans. Circulation 2008; 117: 678–685. Brydon L, O’Donnell K, Wright CE, Wawrzyniak AJ, Wardle J, Steptoe A. Circulating leptin and stress-induced cardiovascular activity in humans. Obesity (Silver Spring) 2008; 16: 2642–2647. Gannage-Yared MH, Khalife S, Semaan M, Fares F, Jambart S, Halaby G. Serum adiponectin and leptin levels in relation to the metabolic syndrome, androgenic profile and somatotropic axis in healthy non-diabetic elderly men. Eur J Endocrinol 2006; 155: 167–176. Hanley AJ, Bowden D, Wagenknecht LE, Balasubramanyam A, Langfeld C, Saad MF et al. Associations of adiponectin with body fat distribution and insulin sensitivity in nondiabetic Hispanics and African-Americans. J Clin Endocrinol Metab 2007; 92: 2665–2671. Hara T, Fujiwara H, Nakao H, Mimura T, Yoshikawa T, Fujimoto S. Body composition is related to increase in plasma adiponectin levels rather than training in young obese men. Eur J Appl Physiol 2005; 94: 520–526. Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y et al. Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol 2000; 20: 1595–1599. Hukshorn CJ, Lindeman JH, Toet KH, Saris WH, Eilers PH, Westerterp-Plantenga MS et al. Leptin and the proinflammatory state associated with human obesity. J Clin Endocrinol Metab 2004; 89: 1773–1778. Nakatani H, Hirose H, Yamamoto Y, Saito I, Itoh H. Significance of leptin and high-molecular weight adiponectin in the general population of Japanese male adolescents. Metabolism 2008; 57: 157–162. Ogawa T, Hirose H, Yamamoto Y, Nishikai K, Miyashita K, Nakamura H et al. Relationships between serum soluble leptin receptor level and serum leptin and adiponectin levels, insulin resistance index, lipid profile, and leptin receptor gene polymorphisms in the Japanese population. Metabolism 2004; 53: 879–885. Plaisance E, Grandjean P, Judd R, Jones K, Taylor J. The influence of sex, body composition, and nonesterified fatty acids on serum adipokine concentrations. Metabolism 2009; 58: 1557–1563. Rush EC, Plank LD, Yajnik CS. Interleukin-6, tumour necrosis factor-alpha and insulin relationships to body composition, metabolism and resting energy expenditure in a migrant Asian Indian population. Clin Endocrinol 2007; 66: 684–690.

International Journal of Obesity

Lack of BMI–adiponectin association S-M Kuo and MM Halpern

1494 56 Hainer V, Hlavata K, Gojova M, Kunesova M, Wagenknecht M, Kopsky V et al. Hormonal and psychobehavioral predictors of weight loss in response to a short-term weight reduction program in obese women. Physiol Res 2008; 57(Suppl 1): S17–S27. 57 Hyatt TC, Phadke RP, Hunter GR, Bush NC, Munoz AJ, Gower BA. Insulin sensitivity in African-American and white women: association with inflammation. Obesity 2009; 17: 276–282. 58 Kondo T, Kobayashi I, Murakami M. Effect of exercise on circulating adipokine levels in obese young women. Endocr J 2006; 53: 189–195. 59 Manning PJ, Sutherland WH, McGrath MM, de Jong SA, Walker RJ, Williams MJ. Postprandial cytokine concentrations and meal composition in obese and lean women. Obesity (Silver Spring) 2008; 16: 2046–2052. 60 Olszanecka-Glinianowicz M, Zahorska-Markiewicz B, Janowska J, Zurakowski A. Serum concentrations of nitric oxide, tumor necrosis factor (TNF)-alpha and TNF soluble receptors in women with overweight and obesity. Metabolism 2004; 53: 1268–1273. 61 Ryan A, Nicklas B, Berman D, Elahi D. Adiponectin levels do not change with moderate dietary induced weight loss and exercise in obese postmenopausal women. Int J Obes Relat Metab Disord 2003; 27: 1066–1071. 62 Valsamakis G, McTernan P, Chetty R, Al Daghri N, Field A, Hanif W et al. Modest weight loss and reduction in waist circumference after medical treatment are associated with favorable changes in serum adipocytokines. Metabolism 2004; 53: 430–434. 63 Trujillo M, Scherer P. AdiponectinFjourney from an adipocyte secretory protein to biomarker of the metabolic syndrome. J Intern Med 2005; 257: 167–175. 64 Schraw T, Wang Z, Halberg N, Hawkins M, Scherer P. Plasma adiponectin complexes have distinct biochemical characteristics. Endocrinology 2008; 149: 2270–2282. 65 Tschritter O, Fritsche A, Thamer C, Haap M, Shirkavand F, Rahe S et al. Plasma adiponectin concentrations predict insulin sensitivity of both glucose and lipid metabolism. Diabetes 2003; 52: 239–243. 66 Tao T, Wickham E, Fan W, Yang J, Liu W. Distribution of adiponectin multimeric forms in Chinese Women with polycystic ovary syndrome and their relation to insulin resistance. Eur J Endocrinol 2010; 163: 399–406. 67 Mullen K, Pritchard J, Ritchie I, Snook L, Chabowski A, Bonen A et al. Adiponectin resistance precedes the accumulation of skeletal muscle lipids and insulin resistance in high-fat-fed rats. Am J Physiol Regul Integr Comp Physiol 2009; 296: R243–R251. 68 Bauche I, Ait El Mkadem S, Rezsohazy R, Funahashi T, Maeda N, Miranda L et al. Adiponectin downregulates its own production and the expression of its adipoR2 receptor in transgenic mice. Biochem Biophys Res Commun 2006; 345: 1414–1424. 69 Behre C. Adiponectin and its role. Scand J Clin Lab Invest 2008; 68: 1–3. 70 Liu M, Liu F. Transcriptional and post-translational regulation of adiponectin. Biochem J 2010; 425: 41–52. 71 Gustafson B, Hammarstedt A, Andersson C, Smith U. Inflamed adipose tissue: a culprit underlying the metabolic syndrome and atherosclerosis. Arterioscler Thromb Vasc Biol 2007; 27: 2276–2283. 72 Sun J, Xu Y, Dai Z, Sun Y. Intermittent high glucose stimulate MCP-l, IL-18, and PAI-1, but inhibit adiponectin expression and secretion in adipocytes dependent of ROS. Cell Biochem Biophys 2009; 55: 173–180. 73 Oller do Nascimento C, Ribeiro E, Oyama L. Metabolism and secretory function of white adipose tissue: effect of dietary fat. An Acad Bras Cienc 2009; 81: 453–466. 74 Lapointe A, Tchernof A, Lamarche B, Piche´ M, Weisnagel J, Bergeron J et al. Plasma adiponectin concentration is strongly associated with VLDL-TG catabolism in postmenopausal women. Nutr Metab Cardiovasc Dis 2010. (E-pub ahead of print). 75 Fasshauer M, Kralisch S, Klier M, Lossner U, Bluher M, Klein J et al. Adiponectin gene expression and secretion is inhibited by

International Journal of Obesity

76 77 78

79 80

81

82

83

84

85

86 87

88

89

90

91

92

93

94 95

interleukin-6 in 3T3-L1 adipocytes. Biochem Biophys Res Commun 2003; 301: 1045–1050. Perez-Martinez P, Perez-Jimenez F, Lopez-Miranda J. n-3 PUFA and lipotoxicity. Biochim Biophys Acta 2010; 1801: 362–366. Mooradian A. Dyslipidemia in type 2 diabetes mellitus. Nat Clin Pract Endocrinol Metab 2009; 5: 150–159. Gade W, Schmit J, Collins M, Gade J. Beyond obesity: the diagnosis and pathophysiology of metabolic syndrome. Clin Lab Sci 2010; 23: 51–61. Cnop M. Fatty acids and glucolipotoxicity in the pathogenesis of type 2 diabetes. Biochem Soc Trans 2008; 36: 348–352. Virtue S, Vidal-Puig A. Adipose tissue expandability, lipotoxicity and the metabolic syndromeFan allostatic perspective. Biochim Biophys Acta 2010; 1801: 338–349. Krogh-Madsen R, Møller K, Dela F, Kronborg G, Jauffred S, Pedersen B. Effect of hyperglycemia and hyperinsulinemia on the response of IL-6, TNF-alpha, and FFAs to low-dose endotoxemia in humans. Am J Physiol Endocrinol Metab 2004; 286: E766–E772. Holman R, Adams C, Nelson R, Grater S, Jaskiewicz J, Johnson S et al. Patients with anorexia nervosa demonstrate deficiencies of selected essential fatty acids, compensatory changes in nonessential fatty acids and decreased fluidity of plasma lipids. J Nutr 1995; 125: 901–907. Mullen K, Tishinsky J, Robinson L, Dyck D. Skeletal muscle inflammation is not responsible for the rapid impairment in adiponectin response with high fat feeding in rats. Am J Physiol Regul Integr Comp Physiol 2010; 299: R500–R508. Gallo S, Egeland G, Meltzer S. Plasma fatty acids are associated with circulating adiponectin in overweight adolescent girls. FASEB J 2007; 21: 834.10. Gallo S, Egeland G, Meltzer S, Legault L, Kubow S. Plasma fatty acids and desaturase activity are associated with circulating adiponectin in healthy adolescent girls. J Clin Endocrinol Metab 2010; 95: 2410–2417. Ferna´ndez-Real J, Vendrell J, Ricart W. Circulating adiponectin and plasma fatty acid profile. Clin Chem 2005; 5: 603–609. Bernsteina E, Koutkiaa P, Ljungquista K, Breua J, Canavana B, Grinspoon S. Acute regulation of adiponectin by free fatty acids. Metabolism 2004; 53: 790–793. Plaisance E, Lukasova M, Offermanns S, Zhang Y, Cao G, Judd R. Niacin stimulates adiponectin secretion through the GPR109A receptor. Am J Physiol Endocrinol Metab 2009; 296: E549–E558. Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K et al. The fat-derived hormone adiponectin reverses insulin resistance associated with both lipoatrophy and obesity. Nat Med 2001; 7: 941–946. Skurk T, Alberti-Huber C, Herder C, Hauner H. Relationship between adipocyte size and adipokine expression and secretion. J Clin Endocrinol Metab 2007; 92: 1023–1033. Pussinen P, Tuomisto K, Jousilahti P, Havulinna AS, Sundvall J, Salomaa V. Endotoxemia, immune response to periodontal pathogens, and systemic inflammation associate with incident cardiovascular disease events. Arterioscler Thromb Vasc Biol 2007; 27: 1433–1439. Radaelli A, Loardi C, Cazzaniga M, Balestri G, DeCarlini C, Cerrito MG et al. Inflammatory activation during coronary artery surgery and its dose-dependent modulation by statin/ACE-inhibitor combination. Arterioscler Thromb Vasc Biol 2007; 27: 2750–2755. Danesh J, Kaptoge S, Mann AG, Sarwar N, Wood A, Angleman S et al. Long-term interleukin-6 levels and subsequent risk of coronary heart disease: two new prospective studies and a systematic review. PLoS Med 2008; 5: e78. O’Rahilly S. Human genetics illuminates the paths to metabolic disease. Nature 2009; 462: 307–314. Shoelson SE, Goldfine AB. Fanning the flames of obesity-induced inflammation. Nat Med 2009; 15: 373–374.