The role of serum adipocyte fatty acid-binding protein on the development of metabolic syndrome is independent of pro-inflammatory cytokines

Methodsand results A total of 465 subjects were selected from participants in a medical check-up programme at a Health Promotion Center. Baseline serum FABP4 levels were measured, and the subjects were evaluated for the presence of metabolic syndrome (MetS) according to the recommendations of the American Heart Association/National Heart, Lung, and Blood Institute. The subjects were re-evaluated 4 years later. Baseline FABP4 concentrations were significantly higher in subjects with MetS than in those without MetS ( P < 0.001). At the 4-year follow-up, subjects in the highest FABP4 tertile at baseline exhibited higher values for body mass index, fat mass and percent body fat, as well as blood pressure, fasting glucose, total cholesterol, triglycerides, low-density lipoprotein (LDL)-cholesterol, insulin, homeostasis model assessment of insulin resistance, monocyte chemoattractant protein-1 and tumor necrosis factor-α levels (all P < 0.05). The subjects with higher FABP4 levels had lower HDL-cholesterol concentrations ( P < 0.05). After adjustment for age, sex, change in percent body fat and baseline values for other metabolic and inflammatory parameters, FABP4 levels at baseline were shown to be strongly associated with the development of MetS by year 4 (odds ratio (OR), 5.75; 95% confidence interval (CI), 2.71–12.23 for highest tertile vs. lowest tertile, P < 0.001) Conclusion Baseline serum FABP4 levels appear to be a significant predictor for the future development of MetS, independent of pro-inflammatory cytokines. Keywords FABP4 Metabolic syndrome Insulin resistance Inflammation Insulin resistance is a hallmark of metabolic syndrome (MetS), a constellation of metabolic abnormalities that includes central obesity, dyslipidaemia, elevated blood pressure and hyperglycaemia [1] . MetS is associated with an increased risk for type 2 diabetes mellitus and cardiovascular disease (CVD) [2,3] . Although the mechanisms underlying MetS are not well understood, current evidence supports the idea that visceral obesity plays an important role [4] . A recent study has reported the dynamic function of adipose tissue as an endocrine organ, not only for storing excess energy, but also for releasing various cytokines, adipokines and inflammatory markers, all of which contribute to the development of insulin resistance and atherosclerosis [5] . Adipocyte fatty acid-binding protein, also known as aP2 or FABP4, is a major cytoplasmic protein that is abundantly expressed in adipocytes and macrophages [6] . FABP4 binds fatty acid ligands with high affinity and functions in intracellular fatty acid trafficking, regulation of lipid metabolism and modulation of gene expression [7,8] . Furthermore, FABP4 deficiency in apoE −/− mice causes a remarkable reduction in atherosclerosis, and improvement in glucose and lipid metabolism [9,10] . Recently, serum FABP4 levels were reported to be associated with the severity of insulin resistance, metabolic syndrome, type 2 diabetes and carotid atherosclerosis in humans [11–13] . These findings suggest that FABP4 plays a key role in the development of MetS and cardiovascular disease, although the precise mechanism is not clear. Based on these findings, we hypothesised that the serum FABP4 level may be a useful maker for identifying patients susceptible to developing MetS, who, in turn, are known to be at risk for cardiovascular events. In the present prospective study, we investigated the relationship between baseline serum FABP4 levels and the development of MetS in apparently healthy adults. To further evaluate the role of FABP4 in obesity-related pathologies, we assessed the relationship between the FABP4 level with components of MetS and various parameters of insulin resistance and inflammation. Methods Participants Among subjects who received annual health check-ups at the Health Promotion Center at Kangbuk Samsung Hospital in Seoul, Korea in 2003, 465 subjects were selected to participate in the baseline study. We excluded patients from the study if they had a medical illness, such as acute infections, chronic renal failure, malignancies or other severe medical illnesses. Written informed consent was obtained from each participant, and the study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected in a priori approval by the institution’s Human Research Committee. This study was approved by the Institutional Review Board at Kangbuk Samsung Hospital. Anthropometric and laboratory measurements The body mass index (BMI; kg/m 2 ) was calculated as the body weight in kilograms divided by the square of height in metres. Systolic and diastolic blood pressures were measured in duplicate and the results were averaged. Fasting blood samples were collected for the measurement of plasma glucose, total cholesterol, triglycerides, high-density lipoprotein-cholesterol (HDL-C) and low-density lipoprotein-cholesterol (LDL-C). The measurement techniques included the hexokinase method for glucose, an enzymatic colourimetric assay for total cholesterol, LDL-C, HDL-C and triglycerides and an immunoradiometric assay for insulin (Biosource, Nivelles, Belgium). Insulin resistance was assessed using the homeostasis model assessment of insulin resistance (HOMA-IR) according to the following equation: fasting blood insulin (uU/ml) × fasting blood glucose (mmol/l)/22.5 [14] . Serum FABP4 concentrations were measured using an enzyme-linked immunosorbent assay (ELISA; BioVendor Laboratory Medicine, Modrice, The Czech Republic). The intra- and inter-assay coefficients of variation (CVs) were 5.0% and 9.8%, respectively. Serum monocyte chemoattractant protein (MCP)-1 and tumor necrosis factor-alpha (TNF-α) were also measured using an ELISA (R&D Systems, Abingdon, UK). Measurement of body fat Total fat mass was determined using tetrapolar bioelectrical impedance analysis (Inbody 2.0; Biospace, Seoul, Korea). Bioelectrical impedance measures two parameters (fat and lean tissue) using empirically derived formulas that have been validated by earlier studies and that correlate well with values obtained using underwater weighing [15] . Diagnosis of diabetes mellitus and MetS The presence of diabetes mellitus was confirmed according to the criteria from the American Diabetes Association [16] or from a participant’s medical history. Based on the American Heart Association/National Heart, Lung, and Blood Institute (AHA/NHLBI) criteria, the diagnosis of MetS was made when the subject satisfied three or more components among the following five components [1] . Because waist circumference measurements were not available for all of the subjects, we substituted BMI (≥25 kg/m 2 ) for waist circumference. (1) obesity: BMI ≥ 25 kg/m 2 ; (2) hypertriglyceridaemia: ≥150 mg/dl (1.7 mmol/l); (3) low HDL-C: <40 mg/dl (1.0 mmol/l) in men and <50 mg/dl (1.3 mmol/l) in women; (4) hypertension: ≥130/85 mmHg; and (5) fasting hyperglycaemia: ≥100 mg/dl (5.6 mmol/l). Statistical analysis Analyses were performed using Statistical Package for Social Sciences (SPSS) for Windows software (version 13.0; SPSS, Chicago, IL, USA). We used an independent sample t -test, analysis of variance, the Mann–Whitney U test and the Kruskal–Wallis test for comparison of continuous variables, and a χ 2 test for comparison of categorical variables. FABP4 levels were grouped into tertiles to simplify the interpretation of the results of subsequent analyses. Serum FABP4 levels, after adjusting for age and sex, were compared according to the number of MetS components present using analysis of covariance. Bivariate correlation analyses between FABP4 and the metabolic parameters were performed using Pearson’s correlation analysis. Multiple logistic regression analysis was used to assess the odds ratio (OR) for the development of MetS by year 4 in subjects in the higher FABP4 tertiles compared with subjects in the lowest tertile, after controlling for confounders in the following steps: (1) age and sex; and (2) change in percent body fat and baseline levels of LDL-C, MCP-1, TNF-α and HOMA-IR, in addition to the variables in step 1. After statistical assessment of the interaction (effect modification between sex and the FABP4 tertiles) based on a multiplicative model by fitting models in step 2, the overall P  value of all of the interaction variables was not significant ( P = 0.961). Thus, we combined both sexes. Fasting blood glucose, triglycerides, fasting insulin, MCP-1, HOMA-IR and FABP4 levels were log-transformed because of a skewed distribution. A P < 0.05 was considered statistically significant. Results The baseline characteristics for all participants are shown in Table 1 . The mean (±SD) age was 40.7 (±6.3) years, and 67% of the participants were male. Among the subjects, 3.4% had diabetes mellitus and 18.5% had MetS. The median (range) for serum FABP4 was 9.58 (7.31–11.85) ng/ml for men and 8.61 (6.72–11.79) ng/ml for women ( P = 0.069). As expected, subjects with MetS at baseline had a greater number of adverse risk factors than subjects without MetS, including a higher LDL-C, insulin resistance index and TNF-α, and a lower HDL-C, as well as the presence of MetS components, such as BMI, fasting glucose, triglycerides, HDL-C and blood pressure ( P < 0.001 for all parameters). The subjects with MetS also had higher FABP4 concentrations ( P < 0.001). Serum log[FABP4] concentrations were correlated with total cholesterol ( r = 0.128, P = 0.007), LDL-C( r = 0.146, P = 0.002), log[MCP-1] ( r = 0133, P = 0.005), TNF-α ( r = 0.318, P < 0.001), fat mass ( r = 0.110, P = 0.019) and percent body fat ( r = 0.126, P = 0.007) after adjusting for age, sex and BMI at baseline. Among the 380 subjects who did not have MetS at baseline, 63 subjects developed MetS after 4 years. Five subjects were on lipid-lowering drugs at year 4, but did not have pre-treatment lipid profiles for accurate classification of the MetS status, and hence were excluded from year 4 analysis. The baseline FABP4 concentrations were significantly higher in subjects who had progressed to MetS by year 4 [8.54 (6.65–11.10) vs. 11.55 (10.00–14.31)] in subjects without MetS ( P < 0.001). With respect to metabolic parameters, subjects in the higher FABP4 tertiles at baseline exhibited higher levels for systolic/diastolic blood pressure, BMI, fasting glucose, total cholesterol, triglycerides, LDL-C, fasting insulin, HOMA-IR, MCP-1 and TNF-α (all P < 0.05) than subjects in the lower tertile after 4 years. In addition, subjects with higher FABP4 levels had lower HDL-C levels. The prevalence of having ≥3 MetS components at year 4 was increased for all baseline FABP4 tertiles ( Table 2 ). Remarkably, serum FABP4 levels increased gradually with an increasing number of MetS components at the 4-year follow-up ( Fig. 1 ). The mean value (standard error (SE)) of FABP4 concentrations for those with 0–5 MetS components was 7.80 (0.09), 8.70 (0.09), 10.47 (0.09), 12.12 (0.20), 12.57 (0.59) and 16.08 (2.36) ng ml −1 , respectively after adjusting for age, sex and BMI. As presented in Table 3 , the baseline serum log[FABP4] levels were strongly correlated with the number of MetS components and TNF-α levels at year 4 after adjusting for age, sex and BMI. Based on multiple regression analysis, baseline log[FABP4] concentrations were significantly associated with the development of MetS after adjusting for age, sex, change in percent body fat and baseline values of LDL-C, MCP-1 and HOMA-IR ( P < 0.01). In addition, patients in the higher tertiles for FABP4 had higher OR for the development of MetS and its components by year 4. In the highest FABP4 tertile compared with the lowest FABP4 tertile, the ORs were 5.75 (95% confidence interval (CI), 2.71–12.23) for MetS, 10.04 (95% CI, 4.85–20.76) for obesity, 3.96 (95% CI, 2.04–7.68) for elevated triglycerides, 2.68 (95% CI, 1.47–4.88) for reduced HDL-C (95% CI, 1.65–4.95), 1.79 (95% CI, 0.93–3.47) for elevated blood pressure and 2.31 (95% CI, 1.24–4.33) for elevated fasting glucose after adjusting for confounding factors ( Table 4 ). After excluding 16 type 2 diabetic patients, the FABP4 levels at baseline were also shown to be strongly associated with the development of MetS by year 4 (OR, 6.37; 95% CI, 2.03–19.98 for the highest tertile vs. the lowest tertile, P < 0.001). After the lower and higher groups were divided into two groups based on the median values, the risk for MetS increased markedly with increasing levels of FABP4, higher levels of MCP-1 ( Fig. 2 (A) ) and higher levels of TNF-α ( Fig. 2 (B)). Moreover, at lower levels of MCP-1 and TNF-α, the risk for MetS was 3–4 fold higher for the higher FABP4 levels. Discussion The current study demonstrated that FABP4 levels are associated with the development of MetS and its components, which are important risk factors for atherosclerosis and type 2 diabetes. Despite the fact that insulin resistance has been considered a major contributor to MetS [17] , we showed that the subjects in the highest FABP4 tertile showed a MetS risk >5-fold higher than those in the lowest tertile, regardless of the degree of HOMA-IR. These results suggest that an elevated baseline FABP4 level may enhance the MetS phenotype via mechanisms other than insulin resistance. The pro-atherogenic activities of FABP4 appear to be mediated by its direct actions in macrophages, through modification of cholesterol trafficking, and through activation of several key inflammatory pathways [18] . Increased systemic inflammation is considered a key factor in the progression of MetS [19,20] , and we hypothesised that FABP4 could be a marker for inflammation and MetS. Recent studies have reported a positive relationship between FABP4 and systemic inflammation reflected by C-reactive protein [12,21] . Furthermore, our study showed that FABP4 was positively associated with inflammatory cytokines, such as MCP-1 and TNF-α, even after adjusting for adiposity. It is possible that FABP4 has a causal role in systemic inflammation, leading to the development of MetS. However, it appears that the relationship between FABP4 and MetS at 4 years was sustained, independent of the presence of an inflammatory marker. The mechanism remains unclear, but these results suggest that excessive production of FABP4 by obese adipose tissue might play a direct role in increasing MetS risk through other mechanisms that do not fully overlap those of the inflammatory pathways. We also found significant relationships between the FABP4 level and the components of MetS. Obesity was more strongly associated with FABP4 than the other components in the present study. Serum FABP4 concentrations were positively correlated with BMI, fat mass and body fat percentage, indicating that adipose tissue is probably a significant contributor of FABP4 into the circulatory system. Therefore, FABP4 might clinically serve as a marker for adiposity, and its possible role as a marker for body composition needs to be explored further. In addition, elevated serum FABP4 levels were associated with increased serum triglycerides and decreased HDL-C levels. These findings are consistent with the findings of a previous study [22] . Recently, Makowski et al. reported that FABP4 altered the cholesterol efflux pathway in macrophages [18] . FABP4 is a critical regulator of the peroxisome proliferator-activated receptors γ-liver X receptor α-adenosine triphosphate (ATP)-binding cassette A1 (ABCA1) pathway and contributes to the control of cholesterol trafficking in macrophages. ABCA1 regulates the rate-limiting step in HDL biogenesis via the efflux of intracellular cholesterol [23] , which might stimulate transfer of triglycerides to HDL and triglycerides catabolism. Thus, this might partly explain the correlation between FABP4, serum triglycerides and HDL-C levels in this study. However, FABP4 is thought to be a cytoplasmic protein; whether or not circulating FABP4 might have an equal influence and pathologic effect at a cellular level requires clarification with further research. Further studies are needed to determine whether or not serum FABP4 plays a causal role in the regulation of lipid metabolism in humans. Baseline FABP4 levels were associated with elevated glucose levels at year 4, even after adjusting for changes in percent body fat and inflammatory cytokines. Previously, Tso et al. reported that elevated baseline serum FABP4 levels predict the development of type 2 diabetes [11] . Although the precise mechanisms explaining the role of FABP4 in glucose metabolism are not yet fully understood, the link between FABP4 and adipose tissue could provide one explanation in glucose metabolism and inflammation [24,25] . Selective deletion of FABP4 in adipocytes resulted in reduced expression of inflammatory cytokines in macrophages, whereas the same deletion in macrophages leads to enhanced insulin signalling and glucose uptake in adipocytes [25] . There is a possibility that FABP4 might affect interactions between adipocytes and macrophages, leading to altered insulin sensitivity and glucose metabolism. There were several limitations to our study. First, the subjects in our study were selected among participants in a health medical check-up and certainly are not representative of the general population due to an underrepresentation of chronic disease. As many as 91% of the baseline participants who qualified for the study were healthier than those populations that do not have access to, or make use of, available medical services. Therefore, the applicability of our study to the general population may be limited. Second, because of the relatively small number of subjects who developed MetS by the 4-year follow-up, additional studies are needed to determine whether or not serum FABP4 levels are associated with the progression of MetS in a larger population. Third, we substituted BMI for waist circumference in the definition of abdominal obesity. However, the accuracy of waist circumference measurements is still a matter of debate. Some investigators have also shown that BMI can predict the development of type 2 diabetes and other metabolic disturbances as robustly as waist circumference [26] . Moreover, the use of BMI has been favourably evaluated as a determinant of MetS. In addition, the conventional BMI calculation could be considered a useful measure of abdominal obesity [20,27] . Finally, certain lifestyle risk factors, such as dietary consumption and exercise habits, which can affect weight gain, were not considered. Thus, the data are subject to potential under- or overestimation. Despite these limitations, our results provide important evidence that the serum FABP4 level can predict the development of MetS on long-term follow-up in apparently healthy Korean adults. Our findings suggest that FABP4 plays a potential role in the pathogenesis of MetS that is independent of changes in adiposity as reflected by percent body fat, insulin resistance as reflected by HOMA-IR and systemic inflammation as reflected by MCP-1 and TNF-α. Further studies are warranted to confirm the significance of FABP4 as a new marker for the progression of MetS in larger, long-term follow-up studies involving more diverse populations. Acknowledgements This work was presented in abstract form at the 22nd Annual Meeting of the Korean Diabetes Association (KDA) in 2009, and supported by a grant awarded by the KDA and Samsung Biomedical Research Institute grant, # SBRI C-A8-223-2 . References [1] S.M. Grundy J.I. Cleeman S.R. Daniels K.A. Donato R.H. Eckel B.A. Franklin Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement Circulation 112 2005 2735 2752 [2] D.E. Laaksonen H.M. Lakka L.K. Niskanen G.A. Kaplan J.T. Salonen T.A. Lakka Metabolic syndrome and development of diabetes mellitus: application and validation of recently suggested definitions of the metabolic syndrome in a prospective cohort study Am J Epidemiol 156 2002 1070 1077 [3] H.M. Lakka D.E. Laaksonen T.A. Lakka L.K. Niskanen E. Kumpusalo J. Tuomilehto The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men JAMA 288 2002 2709 2716 [4] T.S. Han K. Williams N. Sattar K.J. Hunt M.E. Lean S.M. Haffner Analysis of obesity and hyperinsulinemia in the development of metabolic syndrome: San Antonio Heart Study Obes Res 10 2002 923 931 [5] A. Fortuno A. Rodriguez J. Gomez-Ambrosi G. Fruhbeck J. Diez Adipose tissue as an endocrine organ: role of leptin and adiponectin in the pathogenesis of cardiovascular diseases J Physiol Biochem 59 2003 51 60 [6] L. Makowski G.S. Hotamisligil Fatty acid binding proteins – the evolutionary crossroads of inflammatory and metabolic responses J Nutr 134 2004 2464S 2468S [7] N.R. Coe D.A. Bernlohr Physiological properties and functions of intracellular fatty acid-binding proteins Biochim Biophys Acta 1391 1998 287 306 [8] A.V. Hertzel D.A. Bernlohr The mammalian fatty acid-binding protein multigene family: molecular and genetic insights into function Trends Endocrinol Metab 11 2000 175 180 [9] L. Makowski J.B. Boord K. Maeda V.R. Babaev K.T. Uysal M.A. Morgan Lack of macrophage fatty-acid-binding protein aP2 protects mice deficient in apolipoprotein E against atherosclerosis Nat Med 7 2001 699 705 [10] J.B. Boord K. Maeda L. Makowski V.R. Babaev S. Fazio M.F. Linton Combined adipocyte–macrophage fatty acid-binding protein deficiency improves metabolism, atherosclerosis, and survival in apolipoprotein E-deficient mice Circulation 110 2004 1492 1498 [11] A.W. Tso A. Xu P.C. Sham N.M. Wat Y. Wang C.H. Fong Serum adipocyte fatty acid binding protein as a new biomarker predicting the development of type 2 diabetes: a 10-year prospective study in a Chinese cohort Diabetes Care 30 2007 2667 2672 [12] A. Xu A.W. Tso B.M. Cheung Y. Wang N.M. Wat C.H. Fong Circulating adipocyte-fatty acid binding protein levels predict the development of the metabolic syndrome: a 5-year prospective study Circulation 115 2007 1537 1543 [13] D.C. Yeung A. Xu C.W. Cheung N.M. Wat M.H. Yau C.H. Fong Serum adipocyte fatty acid-binding protein levels were independently associated with carotid atherosclerosis Arterioscler Thromb Vasc Biol 27 2007 1796 1802 [14] D.R. Matthews J.P. Hosker A.S. Rudenski B.A. Naylor D.F. Treacher R.C. Turner Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man Diabetologia 28 1985 412 419 [15] C. Vache P. Rousset P. Gachon A.M. Gachon B. Morio A. Boulier Bioelectrical impedance analysis measurements of total body water and extracellular water in healthy elderly subjects Int J Obes Relat Metab Disord 22 1998 537 543 [16] Standards of medical care in diabetes–2007 Diabetes Care 30 Suppl. 1 2007 S4 S41 [17] R.H. Eckel S.M. Grundy P.Z. Zimmet The metabolic syndrome Lancet 365 2005 1415 1428 [18] L. Makowski K.C. Brittingham J.M. Reynolds J. Suttles G.S. Hotamisligil The fatty acid-binding protein, aP2, coordinates macrophage cholesterol trafficking and inflammatory activity. Macrophage expression of aP2 impacts peroxisome proliferator-activated receptor gamma and IkappaB kinase activities J Biol Chem 280 2005 12888 12895 [19] S.M. Grundy Inflammation, hypertension, and the metabolic syndrome JAMA 290 2003 3000 3002 [20] P.M. Ridker J.E. Buring N.R. Cook N. Rifai C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American women Circulation 107 2003 391 397 [21] J.H. Koh Y.G. Shin S.M. Nam M.Y. Lee C.H. Chung J.Y. Shin Serum adipocyte fatty acid-binding protein levels are associated with nonalcoholic fatty liver disease in type 2 diabetic patients Diabetes Care 32 2009 147 152 [22] A. Xu Y. Wang J.Y. Xu D. Stejskal S. Tam J. Zhang Adipocyte fatty acid-binding protein is a plasma biomarker closely associated with obesity and metabolic syndrome Clin Chem 52 2006 405 413 [23] J.F. Oram R.M. Lawn ABCA1. The gatekeeper for eliminating excess tissue cholesterol J Lipid Res 42 2001 1173 1179 [24] K. Maeda H. Cao K. Kono C.Z. Gorgun M. Furuhashi K.T. Uysal Adipocyte/macrophage fatty acid binding proteins control integrated metabolic responses in obesity and diabetes Cell Metab 1 2005 107 119 [25] M. Furuhashi G. Tuncman C.Z. Gorgun L. Makowski G. Atsumi E. Vaillancourt Treatment of diabetes and atherosclerosis by inhibiting fatty-acid-binding protein aP2 Nature 447 2007 959 965 [26] J. Stevens D. Couper J. Pankow A.R. Folsom B.B. Duncan F.J. Nieto Sensitivity and specificity of anthropometrics for the prediction of diabetes in a biracial cohort Obes Res 9 2001 696 705 [27] N. Sattar A. Gaw O. Scherbakova I. Ford D.S. O’Reilly S.M. Haffner Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study Circulation 108 2003 414 419