Correlation of Irisin and Adipokine Levels With Body Mass Index and Risk Factors for Metabolic Syndrome in Hispanic Children

Background Childhood obesity is one of the most serious global public health challenges of the 21st century (Daniels et al., 2009). Mexico has the highest prevalence of obesity, worldwide; 34.4% of children and 35% of adolescents are overweight or obese (ENSANUT 2012). Among other complications, overweight and obese children are predisposed to develop dyslipidemias, hypertension, the metabolic syndrome (MetS) and non-alcoholic fatty liver disease at a younger age, as well as to become obese as adults (Daniels et al., 2009; Elizondo-Montemayor et al., 2010). On the other hand, undernutrition in children conveys to very detrimental health effects such as stunting, puberty retardation, motor retardation (Roulet et al., 2005), osteoporosis and fractures (Saunders et al., 2011), immunologic deficit, and altered function of every organ and system, such as the heart, lungs, kidneys, gastrointestinal tract and the nervous system, among others (Lecours et al, 2001; Saunders et al., 2011). A balance between energy intake and energy expenditure is required to sustain a normal body weight. Anything that disrupts this balance may cause individuals to be underweight, overweight or obese. Diverse signals, central and peripheric participate in the regulation of energy balance. Fat has been considered an endocrine organ due to the many peptides and hormones that it secrets that act upon peripheral and central tissues, and their contribution to insulin resistance or sensitivity and body weight regulation among other functions (Elizondo, 2011). Recently, skeletal muscle has been shown to function as a peripheral endocrine organ by releasing myokines, peptide signals, which are implicated in the regulation of metabolic pathways (Pedersen and Febbraio, 2012). Most recently, a new identified hormone secreted by muscle tissue in mouse, irisin, has been discovered. Irisin acts on white adipose cells in culture and in vivo to stimulate UCP1 expression and a broad program of brown-fat-like development. In mouse, PGC1alfa expression in muscle stimulates increase in expression of FNDC5, a membrane protein that is cleaved and secreted as this novel hormone, irisin. Irisin is induced with exercise in mice and humans, and medley increased irisin levels in blood cause an increase in energy expenditure in mice with no changes in movement or food intake (Boström et al., 2012). Irisin was thus, promptly hypothesized as a hormone influencing body weight, obesity and type 2 diabetes mellitus, among other conditions (Sanchis-Gomar et al., 2012). Some studies have indicated that circulating levels of irisin in humans correlate positively with anthropometric parameters such as BMI, fat mass, fat free mass, and are higher as these parameters increase, this is, irisin levels are higher in obese patients compared to lean ones (Stengel et al., 2013; Huh et al., 2012; Roca-Rivada et al., 2013; Crujeiras et al., 2014; Pardo et al., 2014). Studies have shown an association between irisin levels, insulin resistance and the metabolic syndrome (Park et al., 2013; de la Iglesia et al., 2014; Crujeiras et al., 2014; Pardo et al., 2014). However, some others have found a negative correlation with anthropometric parameters, finding lower irisin levels in obese patients (Moreno-Navarrete et al., 2013). Noteworthy, all these studies have been performed in adults in different circumstances or diseased states. To date, there are only two studies evaluating irisin levels in children. One found that a 1-year long lifestyle intervention program was associated with improvement in anthropometric and metabolic parameters and led to an elevation in irisin levels in obese children, although no correlation was found between irisin levels and anthropometric markers (Blüher et al., 2014). The other study investigated normal weight Saudi children and found correlations between circulating irisin and glucose and HDLc, but a negative association with insulin resistance (Al-Daghri et al., 2014). Besides, associations between irisin levels and adiponectin, leptin and resistin in the set of obesity have been explored. Leptin plays a pivotal role in regulating energy homeostasis, food intake and many neuroendocrine functions, specially triggering puberty in children. Leptin is increased in obese patients, associated with a leptin-resistant state (Blüher and Mantzoros, 2009). Adiponectin is an insulin sensitizing hormone; obese patients have lower levels than normal-weight ones. Circulating adiponectin levels are low in central obesity (Dalamaga et al., 2012), and this low level has been associated with the metabolic syndrome across all ages (Siitonen et al., 2011). Resistin has been associated with insulin resistance, and recently as a proinflamatory adipocytokine (McTernan et al., 2006). Some studies have found a negative correlation between irisin and adiponectin levels (Park et al., 2013), while other have found no association either with leptin or adiponectin (Blüher et al., 2014) here are still conflicting data regarding the association of irisin with anthropometric parameters, obesity and the metabolic syndrome, as well as its 'association with other adipokines, and most important, there is scarce data of these associations in children, the objective of this study will be to correlate the circulating irisin and adipokines levels across a broad spectrum of body mass index ranging from undernourished to obese as well as with insulin resistance and risk factors for the metabolic syndrome in Hispanic children. There is still conflicting data regarding the association of irisin with anthropometric parameters, obesity and the metabolic syndrome, as well as its 'association with other adipokines, and most important, there is scarce data of these associations in children. Therefore, the objective of this study will be to correlate the circulating irisin and adipokines levels across a broad spectrum of body mass index ranging from undernourished to obese as well as with insulin resistance and risk factors for the metabolic syndrome in Hispanic children. Study Population The population was previously described. (Elizondo-Montemayor et al., 2014) An open invitation was made to school-aged children from eight public schools representative of all geographical areas of Monterrey, the second largest city in México. Children that accepted the invitation were randomly selected and screened according to BMI percentiles. The sample size is 40 children. The population will be divided into five groups, 8 per group. Four of the groups will be divided according to the CDC body mass index percentiles: 1.) underweight = \<3percentile; 2.) normal weight = \>3 - \< 85 percentile; 3.) Overweight = \>85 - \< 95 percentile, and 4.) obese = \> 95 percentile. The fifth group will correspond to children with known metabolic syndrome. Signed consent was obtained from both parents/care givers and children. Approvals by the Ethics and Research Committees of the School of Medicine Tecnológico de Monterrey and by the State Health Secretariat, as well as by the Education Authorities were obtained. Participants did not receive gratification during the study. Anthropometric and Clinical Evaluation Anthropometric measurements were performed in all participants at each school. Height was determined to the nearest 0.5 cm (portable Seca® stadiometer, North America) and weight to the nearest 0.1 kg while children wore light clothing, no socks or shoes (TANITA TBF 300® scale, Arlington, Illinois). Percentage of body fat (%BF) was measured by bioimpedance (same TANITA scale). WC was measured to the nearest 0.1 cm at the level of the umbilicus with a flexible fiberglass tape while the subjects were standing, after gently exhaling, and with no clothing on the area. Tricipital skin fold (TSF) was measured using a Lange skinfold caliper. BMI was calculated by weight-kilograms divided by the square of height-meters. Fat mass and fat free mass were calculated according to the specific predetermined formulas. Measurements were performed at the same time each day by the same three trained registered dietitians (RD) in all children to control the inter-observer variability. Blood pressure was measured by the same physician, using an aneroid sphygmomanometer (Welch-Allyn®) following the American Heart Association (AHA) technique; two measurements were obtained while participants were calmed and seated. Laboratory Assessment Blood samples were taken after a 12-hour overnight fast and were kept at 2 to 8°C and centrifuged within the first 3 hours. Serum total cholesterol (TC), HDL-C, low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), and glucose were measured by reflective photometry (Beer-Lambert's law) using an automated analyzer (Architect c8000; Abbott Laboratories, Abbott Park, IL), with an intra- and inter-assay coefficient variation below 4.7%. Plasma samples were then stored at -80°C. Irisin, adiponectin, insulin and leptin will be measured in plasma media using commercial ELISA kits. The assays will be conducted in 96-well microplates according to the manufacturer's instructions (Irisin, adiponectin, insulin and leptin ELISA kit; Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA). Absorbance from each sample will measured in duplicate using a spectrophotometric microplate reader at wavelength of 450 nm (BioTek Instruments, Winooski, VT, USA). Statistical Methods MINITAB version 16 (Minitab Inc., State College, PA, USA) will be used to analyze the differences between anthropometric and biochemical parameters and irisin levels; Microsoft Excel 2007 (Microsoft Corp., Redmond, WA, USA) will be used to incorporate the input of data. The results will be expressed as mean ± standard deviation (s.d.), as absolute number or as percentage (%) and their corresponding 95% confidence intervals (CI). Comparisons between groups for the dependent variables will be made using paired Student´s t-test for means and McNemar test for proportions. For independent variables, the normality will be revised both, graphically and by the Shapiro-Wilk test, and subsequently, the comparisons between groups will be determined using z-test. To assess the association between anthropometric, clinical and biochemical variables and irisin levels, a simple linear regression will be used. The mean comparisons between gender groups will be determined using t-test for independent samples. All tests will be interpreted based on two-tailed hypothesis. The significance level will be set at 0.05 in all cases.