A-FABP and its association with atherogenic risk profile and insulin resistance in young overweight and obese women

A-FABP and its association with atherogenic risk profile and insulin resistance in young overweight and obese women
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  723 ISSN 1752-036310.2217/BMM.13.61 © 2013 Future Medicine Ltd Biomarkers Med.  (2013) 7 (5), 723–730 Research Article  A-FABP and its association with atherogenic risk profile and insulin resistance in young overweight and obese women Aim: We evaluated the association of A-FABP with proatherogenic risk profile and insulin resistance (IR) in young, nondiabetic, overweight/obese women. Materials & methods: Serum A-FABP, high-sensitivity CRP, adiponectin, glucose, insulin, lipids and apolipoproteins were measured in 104 women (aged 20–45 years; BMI ≥ 25 kg/m 2 ) and age-matched healthy controls (n = 76; BMI <25 kg/m 2 ). All patients underwent blood pressure and anthropometric measurements. Results: A-FABP concentration was related to IR, and anthropometric and atherogenic indices. A-FABP was an independent predictor of triglyceride:high-density lipoprotein cholesterol, explaining 42% of its variation in overweight/obese women. At a cutoff level of 16 ng/ml, A-FABP discriminated between controls and overweight/obese (area under curve = 0.96) with high sensitivity and specificity. A-FABP predicted atherogenic risk, with an odds ratio of 11.2 (95% CI: 3.7–34.2), 7.1 (1.9–27.2) and 6.7 (2.6–17.2) for having elevated triglyceride:high-density lipoprotein cholesterol, apoB and CRP, respectively, and IR with an odds ratio 5.6 (1.8–17.2). Conclusion: A-FABP seems to be a valuable predictor of atherogenic risk profile; if elevated it contributes to cardiovascular disease beyond its effect on IR. KEYWORDS: adipocyte n  adiponectin n  cardiovascular risk n  fatty acid-binding protein n  inflammatory factors n  overweight/obesity Obesity, characterized by excess accumulation of adipose tissue, is the most common risk fac-tor for metabolic syndrome (MS), a cluster of abnormalities that includes dyslipidemia, insulin resistance (IR), Type 2 diabetes and hyperten-sion. Adipose tissue plays a central role in the management of systemic energy stores, as well as in many other processes [1] . Obese subjects are more likely to develop glucose intolerance and diabetes, in part owing to the ability of adipose tissue to secrete adipokines, chemokines and free fatty acids into the bloodstream, having a major impact on energy homeostasis and progression from obesity to atherosclerotic diseases, in par-ticular leading to induction of IR and athero-sclerosis [1–3] . Recently, an important molecular pathway that integrates metabolic and inflam-matory responses was suggested that involves the fatty acid-binding proteins (FABPs), which are present in adipocytes and macrophages [3,4] . FABPs belong to a family of cytosolic chaperones that are involved in systemic regulation of lipid and glucose metabolism. A-FABP, also known as aP2 or FABP4, is expressed predominantly in adipose tissue and macrophages. Earlier animal studies showed that mice with A-FABP deficiency are protected from the development of hyperin-sulinemia, hyperglycemia, IR, dyslipidemia and fatty liver disease, in the context of both genetic and dietary obesity [3,5,6] . Some findings suggest that A-FABP may also play a role in the devel-opment of atherosclerotic diseases in humans and that lifestyle changes lower A-FABP plasma concentration in patients with cardiovascular risk [6] . A-FABP has been proposed as a clinical biomarker for atherogenic dyslipidemia, indepen-dent of obesity and IR, in Type 2 diabetics and nonobese patients with familial combined hyper-lipidemia [7,8] , and associated with inflammatory factors related to obesity and MS in nondiabetic morbidly obese women [9] . We aimed to assess the relationship of A-FABP  with atherogenic dyslipidemia, adipose tissue-derived inflammatory factors and IR, as well as the accuracy of A-FABP for predicting athero-genic risk profile in young nondiabetic women  with overweight/obesity. Materials & methods  n Study design & conduction The study group consisted of 104 young pre-menopausal women aged 20–45 years with over- weight/obesity, recruited from the Outpatient Clinic of the City Hospital (Poland). The rea-son they attend the clinic was to decrease weight. None of them were on a weight loss diet at the time of blood collection. The study group was divided into two sub-groups according to BMI: overweight (n = 48; BMI = 25–29.9 kg/m 2 ) and obese women Aneta Mankowska-Cyl* 1 , Magdalena Krintus 1 , Pawel Rajewski 2  & Grazyna Sypniewska 1 1 Department of Laboratory Medicine, Nicolaus Copernicus University in Torun Ludwik Rydygier Collegium Medicum in Bydgoszcz, Poland 2 Department of Internal Diseases, E. Warminski City Hospital, Bydgoszcz, Poland *Author for correspondence: Tel.: +48 52 585 40 46 Fax: +48 52 585 36 03  part of   Research Article Mankowska-Cyl, Krintus, Rajewski & Sypniewska  Biomarkers Med.  (2013) 7 (5) 724 future science group (n = 56; BMI ≥ 30 kg/m 2 ). The control group (BMI <25 kg/m 2 ) included 76 age-matched  women (20–45 years) recruited on a volun-tary basis. Both control and study groups were characterized by normoglycemia (<5.6 mmol/l; fasting glucose was measured twice within a  week). We accepted the following cutoff values for normal lipids, apoB, atherogenic indices and high-sensitivity CRP (hsCRP): total cholesterol (TC) <5.17 mmol/l; high-density lipoprotein (HDL) cholesterol (HDL-C) ≥ 1.29 mmol/l; low-density lipoprotein (LDL)-C <3.36 mmol/l; triglyceride (TG) <1.69 mmol/l; TC:HDL-C <4 [10] ; apoB <0.9 g/l [11] ; TG:HDL-C <1.3 [12] ; and hsCRP <1 mg/l [13] . The study group  was characterized by mild dyslipidemia: TC ≥ 5.17 mmol/l (found in 25% of subjects), HDL-C <1.29 mmol/l (found in 73% of sub- jects) and TG ≥ 1.69 mmol/l (found in 10% of subjects) were found in 25, 73 and 10% of sub- jects, with maximal concentrations of TC and TG of 6.3 and 1.98 mmol/l, respectively, and a minimal value of HDL-C of 0.95 mmol/l. In each subject, bodyweight, height and waist cir-cumference were measured, and BMI was calcu-lated (kg/m 2 ); blood pressure was also examined in a standard manner. Women included in the study had not taken any contraceptives, anti- inflammatories or other medicines known to affect lipid or carbohydrate metabolism. Other exclusion criteria were diagnosis of polycystic ovary syndrome, thyroid disorders or Cushing’s syndrome. The written informed consent from each participant was obtained and the study was approved by the Bioethics Committee at Nico-laus Copernicus University in Torun Collegium Medicum in Bydgoszcz (Poland). n Blood sampling & laboratory analyses Fasting blood was drawn in the early morning (7.00–9.00 am) on the third day of the menstrual cycle. Serum was obtained within less than 1 h to avoid proteolysis, and stored deep-frozen (-80°C) in small aliquots until assayed for no longer than 8 months. Serum was assayed for HDL-C, TGs, TC, apoA-I and apoB, glucose and insulin (ARCHI-TECT ci8200, Abbott Diagnostics, IL, USA). LDL-C, TG:HDL-C (the index of LDL par-ticle size and surrogate for IR), TC:HDL-C, apoB:apoA-I and Homeostasis Model of Assess-ment–Insulin Resistance (HOMA–IR) val-ues were calculated. Cutoff values for elevated apoB:apoA-I ≥ 0.6 [14]  and HOMA–IR ≥ 2.75  were accepted. hsCRP concentration was measured using the BN II System nephelometer (N High Sensitivity CRP; Siemens Healthcare Diagnostics, IL, USA), which provides excel-lent precision with the coefficient of variation reported by the manufacturer of less than 10%. Coefficients of variation for hsCRP estimated in our laboratory were <3.5 and <4.5% for hsCRP concentrations <1 mg/l and >3 mg/l, respectively. Human A-FABP was determined by a sandwich ELISA method for the quantitative measurement of human A-FABP (BioVendor Laboratory Medi-cine Inc., NC, USA). Coefficients of variation for  A-FABP were below 2.6 and <5.1% for A-FABP concentrations <12.5 and >31.1 mg/l, respec-tively. Total adiponectin was assayed by a sand- wich ELISA (DRG MedTek, R&D, Germany) providing excellent precision with the coefficient of variation reported by the manufacturer of less than 7%. The height (cm), weight (kg), and waist and hip circumferences (cm) were measured using standard methods. Waist circumference was measured in a horizontal plane midway between the distance of the superior iliac crest and the lower margin of the last rib. Hip circumference  was taken around the pelvis at the point of maxi-mal protrusion of the buttocks. The systolic and diastolic blood pressures were measured twice according to the standard procedures, in the sit-ting position after at least 5 min rest, by trained personnel with the use of an automatic blood pressure monitor M6 Comfort (HEM-7223-E, OMRON, Poland). Three consecutive readings  were performed, and the average was recorded. The cutoff value of systolic blood pressure was <130 mmHg and <85 mmHg for diastolic blood pressure [15,16] .  n Statistical methods  All data were presented as mean ± standard deviation (Gaussian distribution of results) or median and the 25th and 75th percentile (non-Gaussian distribution). The Student’s t-test and Mann–Whitney U-test were used to compare dif-ferences. Comparison of mean values between the groups was performed by the analysis of vari-ance or Kruskal–Wallis tests. Pearson’s or Spear-man’s correlation tests were used and multiple regression ana lysis was performed. Using Pear-son’s or Spearman’s correlation, the R  2 -coefficient value within the range of 0.4–0.7 was accepted as moderate [17] . In multiple linear regression ana lysis, some models were created based on the assumption that atherogenic indices or athero-genic factors as the dependent variables corre-late with the independent variables. For multiple   A-FABP, atherogenic risk profile & insulin resistance in overweight & obese women Research Article  725 future science group regression ana lysis, the Snedecor F-test was used. Logistic regression was performed to determine associations between baseline parameters and the occurrence of overweight and obesity. To com-pare the diagnostic utility of adiponectin and  A-FABP, receiver-operating characteristic (ROC) curves were constructed (area under the curve [AUC]) and 95% CI, sensitivity and specificity calculated. Odds ratio with 95% CI were also calculated. p < 0.05 was considered statistically significant. Statistical ana lysis was performed using Statistica 10.0 for Windows (StatSoft, OK, USA). Results  n Baseline characteristics & results of basic statistics Baseline characteristics of the obese, overweight and control groups, including anthropo metric and lipid parameters, HOMA–IR, insulin, hsCRP, A-FABP, adiponectin and apolipo protein concentrations, and values of blood pressure constituting major cardiovascular risk factors, are shown in T   ABLE  1 . Women with obesity, compared  with controls, had higher values of anthropo-metric parameters, concentrations of hsCRP and TG, but lower levels of HDL-C. They also presented significantly higher values of athero-genic indices such as TC:HDL-C, apoB:apoA-I and TG:HDL-C, as well as a nearly threefold higher median concentration of A-FABP in com-parison with the control group. Out of biochemi-cal variables, only median A-FABP and hsCRP concentrations as well as TG:HDL-C were sig-nificantly higher in overweight women. On the contrary, median serum adiponectin was signifi-cantly lower only in the obese compared with the control group. In the study group a moderate positive cor-relation between serum A-FABP concentration, BMI and waist circumference was found ( T   ABLE  2 ) . The present study also showed weak but highly significant positive correlation of A-FABP with hsCRP compared with the anti-inflammatory Table 1. Baseline characteristics of the obese, overweight and control group. ParameterGroupp-value Control (BMI <25; n = 76)Overweight (BMI = 25–29.9; n = 48)Obese (BMI ≥ 30; n = 56)Control vs overweight Control vs obeseOverweight vs obese Age (years)28 (25–32)29 (25–36)30 (27–37)NSNSNSWC (cm)71 (69–74)88 (84–90)108 (100–119)0.00010.00010.0001BMI (kg/m 2 )21.4 (19.6–22.3)27.6 (26.3–28.5)35.2 (32.3–38.4)0.00010.00010.0001WHR0.75 (0.73–0.80)0.81 (0.79–0.85)0.87 (0.85–0.91)0.00010.00010.0003A-FABP (ng/ml)10.1 (7.9–13.1)17.5 (14.9–21.7)28 (22.9–42.5)0.00070.00010.03hsCRP (mg/l)0.7 (0.3–1.0)1.4 (0.8–2.5)3.3 (1.9–6.9)0.0050.00010.02Adiponectin (µg/ml)15.5 (9.7–17.9)11.6 (8.5–16.2)10.1 (6.5–13.3)NS0.02NSTC (mmol/l)4.24 (4–4.78)4.11 (3.51–5.01)4.75 (4.16–5.32)NSNSNSLDL-C (mmol/l)2.56 (2.17–2.97)2.4 (1.86–2.76)2.94 (2.43–3.49)NSNS0.02HDL-C (mmol/l)1.39 (1.29–1.57)1.29 (1.11–1.57)1.16 (1–1.29)NS0.00010.03TG (mmol/l)0.72 (0.57–0.82)0.87 (0.65–1.2)1.13 (0.96–1.4)NS0.00010.03TC:HDL-C2.9 (2.7–3.5)3.1 (2.6–3.6)4.4 (3.5–4.9)NS0.00010.0007TG:HDL-C0.48 (0.39–0.57)0.57 (0.52–0.87)1.05 (0.78–1.27)0.010.00010.008apoA-I (g/l)1.5 (1.39–1.6)1.46 (1.22–1.65)1.32 (1.26–1.5)NSNSNSapoB (g/l)0.61 (0.54–0.75)0.64 (0.49–0.76)0.84 (0.7–0.96)NS0.00020.0001apoB:apoA-I0.42 (0.35–0.48)0.43 (0.36–0.53)0.64 (0.49–0.75)NS0.00010.0003Systolic BP (mmHg)120 (114–120)120 (113–128)135 (130–149)NS0.00010.0005Diastolic BP (mmHg)77 (70–80)80 (70–85)90 ( 85–101)NS0.00010.0004Insulin (µIU/ml)5.9 (4.4–7.1)6.0 (4.9–10.7)13.3 (7.8–17.9)NS0.00010.008HOMA–IR1.22 (0.9–1.5)1.32 (0.9–2.4)2.9 (1.8–4.4)NS0.00010.004 Values are presented as median (25th–75th quartile). BP: Blood pressure; HDL-C: High-density lipoprotein cholesterol; HOMA–IR: Homeostasis Model of Assessment–Insulin Resistance; hsCRP: High-sensitivity CRP; LDL-C: Low-density lipoprotein cholesterol; NS: Not statistically significant; TC: Total cholesterol; TG: Triglyceride; WC: Waist circumference; WHR: Waist-to-hip ratio.  Research Article Mankowska-Cyl, Krintus, Rajewski & Sypniewska  Biomarkers Med.  (2013) 7 (5) 726 future science group indicators adiponectin, HDL-C and apoA-I (neg-ative correlation). Moderate positive correlations  were found with HOMA–IR, insulin and indi-ces of atherogenic profile, especially TC:HDL-C, TG:HDL-C and apoB:apoA-I. n Multiple linear & logistic regression ana lysis Furthermore, we performed a multiple linear regression ana lysis. First, we have presented only the results of the best model with two variables explaining changes in TG:HDL-C (the index of LDL particle size) in relation to A-FABP and adiponectin within the study group. This model explained 47% of TG:HDL-C varia-tion (R  2  = 0.47; p = 0.00001); however, only  A-FABP contributed significantly and its impact accounted for 42%.  Another four-variable model with A-FABP as a dependent variable and TG:HDL, HOMA–IR, apoB:apoA1 and hsCRP as independent variables explained 51% of A-FABP variation (R  2  = 0.51; p = 0.00001); however, only TG:HDL and hsCRP contributed significantly. The probability of overweight and obe-sity occurrence, depending on the measured laboratory parameters and calculated atherogenic ratios, was presented using logistic regression after adjustment for age and hypertension. We decided to exclude the BMI from the regression model because of its high colinearity with A-FABP. The designed model was highly significant (p = 0.001) and, based on the result of a pseudomeasure of quality of the fit-R2 Nagelkerke, explained 86% of the variation for overweight and obesity occur-rence. In this model, TC:HDL-C, TG:HDL-C, apoB:apoA-I, HOMA–IR, hsCRP and adipo-nectin did not facilitate the risk stratification for overweight/obesity. Only A-FABP significantly increased the risk of occurrence of overweight and obesity in women by approximately 83% (adjusted odds ratio: 1.83; 95% CI: 1.16–2.89). n Diagnostic utility of investigated markers Finally, we evaluated the ROC curves to assess diagnostic accuracies of investigated variables for the prediction of overweight/obesity occur-rence. The higher level of discrimination of overweight/obesity was found for A-FABP (AUC = 0.96) compared with adiponectin (AUC = 0.67). The best cutoff value of A-FABP for discrimination between normal weight and overweight/obese women was considered to be 16.0 ng/ml (AUC = 0.96; 95% CI: 0.88–0.99; sensitivity 80%; specificity 100%).ROC ana lysis performed to compare A-FABP and adiponectin discriminating power for pre-dicting an atherogenic risk profile has shown better accuracy for A-FABP ( T   ABLE  3 ) . The diag-nostic accuracy of A-FABP for predicting elevated TG:HDL-C in overweight/obese women was bet-ter compared with that of adiponectin. Similarly, significantly higher diagnostic accuracy was found for A-FABP as a predictor of increased hsCRP concentration in overweight/obese women than for adiponectin. For all parameters, sensitivity and specificity were calculated as shown in T   ABLE  3 .Serum A-FABP concentration over 16 ng/ml was associated with an odds ratio of 11.2 (p = 0.0001), 7.1 (p = 0.004), 6.7 (p = 0.0001) and 5.1 (p = 0.002) for elevated TG:HDL-C, apoB, hsCRP or apoB:apoA-I, respectively ( T   ABLE  4 ) . Moreover, A-FABP concentrations over the cutoff predicted IR (HOMA–IR) with an odds ratio of 5.6 (p = 0.002). Discussion Subjects with obesity and diabetes mellitus are at increased risk for cardiovascular diseases. Cur-rently, the diagnosis of cardiovascular diseases in these asymptomatic patients is not adequate, and Table 2. Spearman’s correlation coefficients between A-FABP and the measured parameters in the study group. ParameterA-FABP R  2  p-value hsCRP 0.340.001Adiponectin-0.310.03TG0.290.03LDL-CNSNSHDL-C-0.300.04TC:HDL-C0.430.004TG:HDL-C0.410.003apoA-I-0.340.02apoB0.300.04apoB:apoA-I0.400.007Systolic BP0.280.05Diastolic BPNSNSInsulin0.410.005HOMA–IR0.430.003Glucose0.320.03BMI0.620.0007WC0.470.001 BP: Blood pressure; HDL-C: High-density lipoprotein cholesterol; HOMA–IR: Homeostasis Model of  Assessment–Insulin Resistance; hsCRP: High-sensitivity CRP; LDL-C: Low-density lipoprotein cholesterol; NS: Not statistically significant; TC: Total cholesterol; TG: Triglyceride; WC: Waist circumference.   A-FABP, atherogenic risk profile & insulin resistance in overweight & obese women Research Article  727 future science group searching for new potential biomarkers is of great importance. In the present study, we have investi-gated the relationship of serum A-FABP, the adi-pose tissue-derived molecule, with atherogenic risk profile and insulin sensitivity in non diabetic young women with overweight/obesity. The asso-ciation of A-FABP with HOMA–IR and proin-flammatory cytokines was found very recently in nondiabetic, morbidly obese Spanish women [9] . We have found that in overweight and mildly obese young women, A-FABP correlated with pro-inflammatory and anti-inflammatory factors and IR indices. A-FABP was an independent predictor of TG:HDL-C (the index of LDL particle size),  whereas neither BMI nor adiponectin have shown a predictive power in this setting. Serum A-FABP  was found to explain 42% of TG:HDL-C varia-tion in overweight/obese women, which indicates its association with the LDL particle size and possibly their number, as well as reflecting IR [18] . According to the previous data, of two main adipose tissue-derived molecules, adiponectin  was considered an inhibitor of atherosclerosis, preventing the onset and progression of cardio-vascular diseases, especially in cases with MS [2,19–21] , whereas A-FABP was associated with carotid atherosclerosis in humans. In contrast to adiponectin, A-FABP promotes inflammation [5,6]  and is proposed to contribute to dyslipidemia and foam cell formation [22,23] . A-FABP might serve as a significant mediator that links obesity  with its related metabolic and cardiovascular diseases [6,24–27] . A-FABP-deficient mice are pro-tected from IR, hyperglycemia and atherosclerosis [28,29] . In humans, it has been demonstrated that the circulating A-FABP level has been associ-ated with central adiposity, IR and subclinical athero sclerosis [3,9,28] . Despite the observation that A-FABP seems to be closely associated to adiposity and fat distribution, it was postulated that lifestyle changes lower plasma A-FABP con-centration in patients with cardiovascular risk [6] . However, data on the impact of lifestyle changes on A-FABP level are scarce.Good positive association between serum  A-FABP concentration and indicators of adipos-ity (BMI and waist circumference) suggests that adipose tissue might be the major contributor of  A-FABP secreted into the circulation. In fact,  A-FABP mRNA expression in visceral adipose tissue was related to its circulating concentrations in morbidly obese women [9] . Data of Tuncman et al.  have provided support to the recently pub-lished study, demonstrating that a genetically determined reduction in the A-FABP mRNA level in human adipose tissue is associated with a reduced risk of cardiovascular disease and sug-gest a pathogenetic role of A-FABP in cardiovas-cular risk in humans, as seen in experimental animals [30] . Table 3. Diagnostic utility of A-FABP and adiponectin for predicting atherogenic risk profile. Atherogenic risk profileParameterAUC (95% CI)p-valueSensitivity/specificity (%) LDL-C ≥ 3.36 mmol/lA-FABP0.63 (0.52–0.74)NS100/27Adiponectin0.63 (0.51–0.73)65/63apoB ≥ 0.9 g/lA-FABP0.73 (0.62–0.83)NS87/58Adiponectin0.65 (0.53–0.75)47/80apoB:apoA1 ≥ 0.6A-FABP0.74 (0.63–0.83)NS75/67Adiponectin0.69 (0.59–0.79)65/72TG:HDL-C ≥ 1.3A-FABP0.88 (0.79–0.94)0.0878/70Adiponectin0.73 (0.62–0.83)85/62TC:HDL-C ≥ 4A-FABP0.77 (0.66–0.86)NS75/70Adiponectin0.70 (0.59–0.79)83/59hsCRP ≥ 1 mg/lA-FABP0.80 (0.70–0.88)0.00384/65Adiponectin0.59 (0.48–0.70)55/66HOMA–IR ≥ 2.75A-FABP0.78 (0.67–0.86)NS83/72Adiponectin0.79 (0.69–0.87)63/92  AUC: Area under the curve; HDL-C: High-density lipoprotein cholesterol; HOMA–IR: Homeostasis Model of  Assessment–Insulin Resistance; hsCRP: High-sensitivity CRP; LDL-C: Low-density lipoprotein cholesterol; NS: Not  statistically significant; TC: Total cholesterol; TG: Triglyceride.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks