Role of the Polymorphisms of Uncoupling Protein Genes in Childhood Obesity and Their Association with Obesity-Related Disturbances

Abstract

Background:

Obesity, one of the most common disorders observed in clinical practice, has been associated with energy metabolism-related protein genes such as uncoupling proteins (UCPs). Herein, we evaluated UCPs as candidate genes for obesity and its morbidities.

Methods:

A total of 268 obese and 185 nonobese children and adolescents were enrolled in this study. To determine dyslipidemia, hypertension, and insulin resistance, laboratory tests were derived from fasting blood samples. UCP1-3826 A/G, UCP2 exon 8 deletion/insertion (del/ins), and UCP3-55C/T variants were also genotyped, and the relationships among the polymorphisms of these UCPs and obesity morbidities were investigated.

Results:

The mean ages of the obese and control groups were 11.61 ± 2.83 and 10.74 ± 3.36 years, respectively. The respective genotypic frequencies of the AA, AG, and GG genotypes of UCP1 were 46.3%, 33.2%, and 20.5% in obese subjects and 46.5%, 42.2%, and 11.4% in the controls (p = 0.020). G alleles were more frequent in obese subjects with hypertriglyceridemia (42.9%; p = 0.048) than in those without, and the GG genotype presented an odds ratio for obesity of 2.02 (1.17-3.47; p = 0.010). The polymorphisms of UCP2 exon 8 del/ins and UCP3-55C/T did not influence obesity risk (p > 0.05). The I (ins) allele was associated with low HDL cholesterolemia (p = 0.023).

Conclusion:

The GG genotype of the UCP1-3826 A/G polymorphism appears to contribute to the onset of childhood obesity in Turkish children. The GG genotype of UCP1, together with the del/del genotype of the UCP2 polymorphism, may increase the risk of obesity with synergistic effects. The ins allele of the UCP2 exon 8 del/ins polymorphism may contribute to low HDL cholesterolemia.

Introduction

The incidence rate of obesity, one of the most common disorders observed in clinical practice, continues to increase dramatically every year (Lobstein et al., 2003). Our knowledge of human obesity has progressed beyond the simple generalization that obesity is fully explained by inappropriate eating. Indeed, excessive fat accumulation in adipose tissue leading to obesity is the result of chronic overconsumption of foods and beverages over the energy expenditure requirement; thus, obesity is a disease in which dietary and lifestyle habits, sociologic factors, metabolic and neuroendocrine alterations, and hereditary components are involved (Marti et al., 2004). About 65% of the variation in obesity is genetic (Brondani et al., 2012) and associated with energy metabolism-related protein genes such as uncoupling proteins (UCPs) (Brondani et al., 2012).

Because the genome-wide association studies have already identified single-nucleotide polymorphisms related to several complex conditions, great efforts have been made to identify genes associated with several disorders, and many studies have focused on genes related to energy expenditure, such as those encoding adrenergic receptors and mitochondrial UCPs (Silva and Rabelo, 1997; Valve et al., 1998; Verdi et al., 2015).

UCPs are believed to transport fatty acid anions from the inner surface of the inner mitochondrial membrane to its outer surface (Garlid et al., 1996). UCP1 plays important roles in the metabolic and energy balance and regulation, cold- and diet-induced thermogenesis, and decreasing the production of reactive oxygen species by mitochondria, all of which are mechanisms associated with the pathogenesis of obesity and/or type 2 diabetes mellitus (DM2) (Dalgaard and Pedersen, 2001). UCP1 plays an important thermogenic role in brown adipose tissue by virtue of its capacity to uncouple mitochondrial respiration from oxidative phosphorylation in a regulated manner (Silva and Rabelo 1997). UCP2 and UCP3 are located within 150 kb of each other on chromosome 11q13 (Pecqueur et al., 1999). Activation of UCP2 may attenuate glucose-stimulated insulin secretion by pancreatic β cells (Rutter, 2001). UCP3 is predominantly expressed in skeletal muscles. Because of its suspected function, UCP3 is one of the many genes that may be associated with early obesity or its effects (Boss et al., 1997).

Some well-known polymorphisms of UCPs have been reported to be functional variants on gene expression: the UCP1-3826 A/G polymorphism has been associated with reduced mRNA expression (Csernus et al., 2013), and a relationship between A to G substitution of the UCP1 gene and an increased capacity to gain weight (Valve et al., 1998; Otabe et al., 2000; Csernus et al., 2013) and resistance to a low-calorie diet have been observed (Valve et al., 1998). The UCP2 45 bp exon 8 deletion/insertion (del/ins) polymorphism has been associated with mRNA processing or transcript stability (Jun et al., 2009). Schrauwen et al. (1999) reported that Pima Indians carrying the −55T allele present a significantly higher UCP3 mRNA concentration than −55C homozygotes, thereby suggesting that the −55T allele can increase UCP3 mRNA expression. The UCP3-55T variant has shown a protective role against fat accumulation, especially in children with low levels of physical activity (Walder et al., 1998).

In this study, UCPs were examined as candidate genes for obesity and were also investigated for association with obesity-associated disturbances such as hypertension, dyslipidemia, and insulin resistance.

Materials and Methods

Participants and study area

A total of 268 obese children and adolescents who were followed at the Hospital of the School of Medicine at Gaziosmanpasa University, Tokat, Turkey, were enrolled in this study. Another 185 nonobese school-aged children from outpatient of pediatric clinics were recruited for the control group. The study protocol was in accordance with the Helsinki Declaration of the World Medical Association and ethical standards, and approval of this study was received from the Ethics Committee of Gaziosmanpasa University School of Medicine (14-KAEK-204). Informed consent was obtained and questionnaires used to gather information were answered by parents and the participants themselves.

Weights were measured using a digital scale (Seca Corp., Chino, CA) and taken while the patients were barefoot and wearing light clothing. Height was measured using a portable stadiometer (Seca Corp.). Body mass index (BMI) was calculated as weight in kilograms divided by height in square meters (kg/m²). The subjects were diagnosed as obese according to a BMI of >95 percentile considering the sex- and age-specific growth curves and cutoff levels proposed by Neyzi et al. (2008).

Laboratory tests

Laboratory tests, including glucose, insulin levels, and lipid profiles, were derived from fasting blood samples drawn only from obese participants. Blood samples were also subjected to genetic analysis for UCP polymorphisms. Besides genetic analyzes, the control group was not subjected to laboratory tests.

Serum fasting glucose, insulin, triglycerides (TGs), and high-density lipoprotein cholesterol (HDL-C) were detected using reagent kits from Roche Diagnostics adapted to the COBAS 6000 Autoanalyzer (Roche Diagnostics, Indianapolis, IN).

Definition of hypertension, insulin resistance, and dyslipidemia

Blood pressure (BP) was measured using a digital sphygmomanometer (OMRON 705IT; Omron Healthcare Co., Kyoto, Japan). If the measured BP was high according to age, sex, and height, the mean of two measurements was noted. Hypertension was defined as a BP ≥95th percentile according to age, sex, and height (Falkner and Daniels, 2004).

The homeostasis model assessment of insulin resistance index (HOMA-IR) was calculated using the following equation: HOMA-IR = fasting insulin (μU/mL) × fasting glucose (mg/dL)/405). The HOMA cutoff for insulin resistance diagnosis was 3.16 (Keskin et al., 2005).

Dyslipidemia was defined as the existence of any of these criteria: TGs >105 mg/dL in children <10 years of age and >136 mg/dL in children ≥10 years of age and HDL-C <35 mg/dL (Sangun et al., 2011).

Genetic analysis

Blood samples were collected from each subject, and DNA was extracted from peripheral blood samples using a GeneAll^® ExgeneTM Blood SV Genomic DNA Kit according to the manufacturer's instructions. UCP1-3826 A/G, UCP2 exon 8 del/ins, and UCP3-55C/T gene polymorphisms were analyzed by a polymerase chain reaction (PCR)-based restriction fragment length polymorphism (RFLP) method. PCR was performed in a total volume of 25 μL containing 25-50 ng of genomic DNA, 0.8 nmol/μL of each primer, 1.5 mM MgCl₂, 2.5 μL of 10 × PCR buffer, 0.3 mM dNTP, and 1 U Taq DNA polymerase (Fermentas, Shenzhen, China). The PCR primers, PCR programs, product sizes, and restriction enzymes are shown in Table 1. The amplified products were run on 3% agarose gel, stained with ethidium bromide, and visualized under ultraviolet light.

Table 1.

Polymerase Chain Reaction Primers, Polymerase Chain Reaction Programs, Product Sizes, and Restriction Enzymes for the UCP1-3826A/G, UCP2 Exon 8 Del/Ins, and UCP3-55C/T Polymorphisms

Statistical analysis

Data are expressed as mean ± standard deviation or frequency and percent. Independent samples t-test was used to compare continuous normal data between groups. The χ² test was used to compare categorical data between/among groups and evaluate the Hardy-Weinberg equilibrium of the distribution of genotypes in patients and controls. A p < 0.05 was considered significant. Analyses were performed using SPSS 19 (IBM SPSS Statistics 19; SPSS, Inc., Somers, NY).

Results

The UCP1-3826 A/G, UCP2 exon 8 del/ins, and UCP3-55C/T variants were genotyped, and associations between the UCP1, UCP2, or UCP3 polymorphisms and BMI, hypertension according to systolic/diastolic systemic BP, and obesity-related biochemical traits (e.g., HDL-C, TG, fasting glucose, insulin, HOMA-IR) in obese children and adolescents were examined. The clinical and laboratory characteristics of the study participants are shown in Table 2.

Table 2.

Clinical and Laboratory Characteristics of Study Participants

	Obese (n = 268)	Control (n = 185)
Age (year)	11.61 ± 2.83	10.74 ± 3.36
Gender (F/M)	163/105	101/84
Length (cm)	150.76 ± 15.17	143.56 ± 17.84
Weight (kg)	65.9 ± 19.96	38.99 ± 14.8
BMI (kg/m²)	27.93 ± 4.39	18.1 ± 3.18
HDL-C (mg/dL)	48.19 ± 12.41
TGs (mg/dL)	111.29 ± 64.42
Insulin (uIU/mL)	19.08 ± 12.39
Glucose (mg/dL)	88.21 ± 12.01
HOMA-IR	4.13 ± 3.09
Hypertension	93 (40.3)
IR	142 (56.3)
Low HDL-C	27 (10.4)
Hypertriglyceridemia	113 (44.0)

Data are shown as mean ± standard deviation or frequency, percentage.

BMI, body mass index; HDL-C, high-density lipoprotein cholesterol, HOMA-IR, the homeostasis model assessment of insulin resistance index; IR, insulin resistance; TGs, triglycerides.

The respective genotype frequencies of the AA, AG, and GG genotypes of UCP1 were 46.3%, 33.2%, and 20.5% in obese subjects and 46.5%, 42.2%, and 11.4% in control subjects (p = 0.020), and the allelic frequency of the G allele was 37.1% in obese subjects and 32.4% in controls (p = 0.146). The genotype frequencies observed were in Hardy-Weinberg equilibrium. G alleles were more frequent (42.9%) in obese subjects with hypertriglyceridemia than in obese subjects with normal levels of TGs (34.4%; p = 0.048). Table 3 shows the clinical characteristics of the study participants according to the genotypes of the UCP1, UCP2, and UCP3 polymorphisms.

Table 3.

Clinical Characteristics of Study Participants According to the Genotypes of UCP1, UCP2, and UCP3 Polymorphisms

	UCP1					UCP2					UCP3
	AA	AG	GG	A	G	DD	ID	II	D	I	CC	CT	C	T
Group
Obese	124 (46.3)	89 (33.2)	55 (20.5)	337 (62.9)	199 (37.1)	151 (56.4)	96 (35.8)	21 (7.8)	398 (74.3)	138 (25.7)	195 (72.8)	73 (27.2)	463 (86.4)	73 (13.6)
Control	86 (46.5)	78 (42.2)	21 (11.4)	250 (67.6)	120 (32.4)	93 (50.2)	68 (36.8)	24 (13)	254 (68.6)	116 (31.4)	148 (80)	37 (20)	333 (90)	37 (10)
χ²; p	7.868; 0.020			2.115; 0.146		3.684; 0.159			3.409; 0.065		3.119; 0.077		2.688; 0.101
Hypertension
No	61 (44.2)	47 (34.1)	30 (21.7)	169 (61.2)	107 (38.8)	81 (58.7)	46 (33.3)	11 (8)	208 (75.4)	68 (24.6)	93 (67.4)	45 (32.6)	231 (83.7)	45 (16.3)
Yes	48 (51.6)	27 (29)	18 (19.4)	123 (66.1)	63 (33.9)	45 (48.4)	40 (43)	8 (8.6)	130 (69.9)	56 (30.1)	74 (79.6)	19 (20.4)	167 (89.8)	19 (10.2)
χ²; p	1.237; 0.539			1.146; 0.284		0.646; 0.724			1.693; 0.193		4.114; 0.043		3.453; 0.006
IR
No	53 (48.2)	37 (33.6)	20 (18.2)	143 (65)	77 (35)	62 (56.4)	37 (33.6)	11 (10)	161 (73.2)	59 (26.8)	78 (70.9)	32 (29.1)	188 (85.5)	32 (14.5)
Yes	63 (44.4)	45 (31.7)	34 (23.9)	171 (60.2)	113 (39.8)	80 (56.3)	52 (36.6)	10 (7)	212 (74.6)	72 (25.4)	104 (73.2)	38 (26.8)	246 (86.6)	38 (13.4)
χ²; p	1.229; 0.541			1.21; 0.271		1.100; 0.577			0.139; 0.710		0.168; 0.682		0.141; 0.708
Low HDL-C
No	103 (44.2)	83 (35.6)	47 (20.2)	289 (62)	177 (38)	126 (54.1)	87 (37.3)	20 (8.6)	339 (72.7)	127 (27.3)	173 (74.2)	60 (25.8)	406 (87.1)	60 (12.9)
Yes	15 (55.6)	4 (14.8)	8 (29.6)	34 (63)	20 (37)	21 (77.8)	5 (18.5)	1 (3.7)	47 (87)	7 (13)	16 (59.3)	11 (40.7)	43 (79.6)	11 (20.4)
χ²; p	4.841; 0.089			0.018; 0.892		5.539; 0.063			5.166; 0.023		2.739; 0.098		2.306; 0.129
High TGs
No	70 (48.6)	49 (34)	25 (17.4)	189 (65.6)	99 (34.4)	82 (56.9)	49 (34)	13 (9)	213 (74)	75 (26)	104 (72.2)	40 (27.8)	248 (86.1)	40 (13.9)
Yes	46 (40.7)	37 (32.7)	30 (26.5)	129 (57.1)	97 (42.9)	62 (54.9)	43 (38.1)	8 (7.1)	167 (73.9)	59 (26.1)	85 (75.2)	28 (24.8)	198 (87.6)	28 (12.4)
χ²; p	3.405; 0.182			3.92; 0.048		0.629; 0.730			0.001; 0.987		0.293; 0.589		0.248; 0.619

Data are shown as n (%).

Statistical significance is shown in bold.

D, deletion (del); DD, del/del; I, insertion (ins); II, ins/ins; IR, insulin resistance.

Homozygotes for the UCP1 GG genotype showed an odds ratio (OR) for obesity of 2.02 (1.17-3.47; p = 0.010; Fig. 1). The BMI levels of the GG genotype (25.56 ± 5.38) were significantly higher than those found in subjects with other (AA and AG) genotypes (23.83 ± 6.34; p = 0.028). By contrast, HDL-C levels were significantly lower for carriers of the GG genotype (44.66 ± 13.72) than for carriers of other genotypes (AG and AA, 49.14 ± 11.89; p = 0.017). The means of quantitative variables are indicated in Table 4.

FIG. 1.

Comparisons of percentages of genotypes of UCPs in obese and control participants. GG homozygous genotype with AG and AA genotypes for UCP1 (χ² = 6.593; p = 0.010; odds ratio [95% CI] = 2.02 [1.17-3.47]). II homozygous genotype with ID and DD genotypes for UCP2 (p > 0.05). CC homozygous genotype with CT heterozygotes for UCP3 (χ² = 3.119; p = 0.077). D, deletion (del); DD, del/del; I, insertion (ins); ID, ins/del; II, ins/ins; UCP, uncoupling protein.

Table 4.

Means of Quantitative Variables Based on Uncoupling Protein Genotypes

	BMI	HOMA-IR	HDL-C (mg/dL)	TGs (mg/dL)	Insulin (uIU/mL)	Glucose (mg/dL)
UCP1
AA	24.35 ± 6.34^ab	3.94 ± 2.74	48.44 ± 12.67^ab	111.97 ± 57.93	18.35 ± 11.12	87.18 ± 14.36
AG	23.19 ± 6.3^b	4.18 ± 3.64	50.1 ± 10.74^b	105 ± 54.08	19.56 ± 14.3	87.92 ± 9.96
GG	25.56 ± 5.38^a	4.46 ± 2.91	44.66 ± 13.72^a	119.7 ± 88.24	19.95 ± 12.02	90.96 ± 8.62
p	0.019	0.582	0.037	0.414	0.672	0.147
UCP1
AA	24.35 ± 6.34	3.94 ± 2.74	48.44 ± 12.67	111.97 ± 57.93	18.35 ± 11.12	87.18 ± 14.36
AG +GG	23.95 ± 6.11	4.29 ± 3.36	47.99 ± 12.23	110.73 ± 69.5	19.71 ± 13.4	89.08 ± 9.56
p	0.503	0.366	0.775	0.879	0.382	0.200
UCP1
GG	25.56 ± 5.38	4.46 ± 2.91	44.66 ± 13.72	119.7 ± 88.24	19.95 ± 12.02	90.96 ± 8.62
AG +AA	23.83 ± 6.34	4.04 ± 3.14	49.14 ± 11.89	109 ± 56.3	18.84 ± 12.5	87.49 ± 12.67
p	0.028	0.374	0.017	0.276	0.560	0.056
UCP2
DD	24.69 ± 6.3^a	4.2 ± 3.4	47.97 ± 13.27	115.05 ± 76.77	19.36 ± 13.49	87.88 ± 12.66
ID	23.85 ± 6.19^ab	4.22 ± 2.82	48.11 ± 11.34	109.16 ± 45.62	18.9 ± 11	89.09 ± 9.21
II	22.17 ± 5.5^b	3.3 ± 1.58	50.1 ± 10.98	94.84 ± 33.18	17.95 ± 10.3	86.6 ± 17.69
p	0.036	0.438	0.762	0.377	0.875	0.606
UCP2
DD	24.69 ± 6.3	4.2 ± 3.4	47.97 ± 13.27	115.05 ± 76.77	19.36 ± 13.49	87.88 ± 12.66
ID+II	23.49 ± 6.08	4.05 ± 2.65	48.48 ± 11.25	106.5 ± 43.81	18.72 ± 10.83	88.64 ± 11.17
p	0.043	0.698	0.742	0.292	0.683	0.608
UCP2
II	22.17 ± 5.5	3.3 ± 1.58	50.1 ± 10.98	94.84 ± 33.18	17.95 ± 10.3	86.6 ± 17.69
ID+DD	24.35 ± 6.26	4.21 ± 3.18	48.03 ± 12.54	112.76 ± 66.34	19.18 ± 12.57	88.35 ± 11.44
p	0.027	0.199	0.464	0.223	0.662	0.522
UCP3
CC	24.05 ± 6.32	4.13 ± 3.2	48.73 ± 11.9	106.87 ± 50.69	19.09 ± 12.76	88.29 ± 10.98
CT	24.37 ± 5.93	4.13 ± 2.81	46.78 ± 13.66	123.59 ± 91.85	19.06 ± 11.46	87.98 ± 14.48
p	0.641	0.997	0.261	0.066	0.987	0.852

Data are shown as mean ± standard deviation.

Statistical significance is shown in bold.

Different superscript letters indicate a statistical significant difference.

UCP2 exon 8 del/ins polymorphisms did not influence the risk of obesity (p > 0.05; Table 3), and homozygotes of ins/ins were associated with lower BMI levels compared with del/del (DD) homozygotes (p = 0.036; Table 3). The ins allele was associated with low HDL cholesterolemia (p = 0.023; Table 3).

While the UCP3-55C/T polymorphisms were not associated with any risk for obesity (p = 0.077), the CC homozygote genotype and C allele were indicated as factors for hypertension in obese subjects (p = 0.043, OR = 1.88 [1.02-3.49] and p = 0.006, OR = 1.71 [0.96-3.03], respectively; Table 3).

To better understand the possible synergistic implication of UCP1 and UCP2 in the development of human obesity, we have divided the study participants into two groups: the first group included subjects who were carriers of both the GG homozygote of the UCP1 polymorphism and the DD homozygote of the UCP2 polymorphism (group 1) and the second group included all other participants (group 2). Group 1 generally showed an OR of 2.69 (p = 0.014) for obesity, and obese subjects in this group demonstrated an OR of 3.57 (p = 0.010) for low HDL cholesterolemia. Table 5 summarizes the clinical characteristics of participants who had both the genotypes, GG genotype of the UCP1 polymorphism and the DD genotype of the UCP2 polymorphism.

Table 5.

Clinical Characteristics of Participants and Obese Subjects with the GG Genotype of the UCP1 Polymorphism and the Del/Del Genotype of the UCP2 Polymorphism

	GG genotype of UCP1 and also DD genotype of UCP2 polymorphisms
Variables	Carriers of both GG and DD genotypes	Subjects except the first group	p	Odds ratio (95% CI)
Group
Obese	29 (10.8)	239 (89.2)	0.014	2.69 (1.20-6.01)
Control	8 (4.3)	177 (95.7)	0.014	2.69 (1.20-6.01)
Hypertension
No	18 (72)	120 (58.3)	0.186	0.54(0.22-1.36)
Yes	7 (28)	86 (41.7)	0.186	0.54(0.22-1.36)
IR
No	10 (35.7)	100 (44.6)	0.369	1.45(0.64-3.28)
Yes	18 (64.3)	124 (55.4)	0.369	1.45(0.64-3.28)
Low HDL-C
No	22 (75.9)	211 (91.3)	0.010	3.57(1.28-8.82)
Yes	7 (24.1)	20 (8.7)	0.010	3.57(1.28-8.82)
High TGs
No	15 (51.7)	129 (56.6)	0.620	1.22(0.56-2.64)
Yes	14 (48.3)	99 (43.4)	0.620	1.22(0.56-2.64)

Data are shown as n (%).

Statistical significance is shown in bold.

Discussion

Considering its morbidities, obesity is an important health problem among children. The condition has been suggested to result not only from excess food intake but also from low energy expenditure. When hereditary properties are considered, polymorphisms of genes related to energy metabolism are believed to promote increased susceptibility to developing obesity.

The results of this study support an association between UCP polymorphisms and obesity and low HDL cholesterolemia because GG homozygous genotype carriers showed a twofold higher risk of developing obesity than noncarriers. Because of the role of UCPs in energy expenditure, we suggest that childhood-onset obesity is associated with UCP polymorphisms.

The GG genotype of the UCP1-3826 A/G polymorphism may independently exert a significant effect on obesity development, and GG genotype individuals indicated twofold higher risks of developing obesity than individuals with other genotypes of UCP1-3826 A/G polymorphisms. Associations between UCP1-3826 A/G polymorphisms and obesity-related phenotypes and phenomena, such as BMI (Matsushita et al., 2003; Nakano et al., 2006), lipid/lipoprotein (Kotani et al., 2008), and DM2 (Sramkova et al., 2007), have been described. Nakano et al. (2006) reported that A/G heterozygotes present higher BMIs than A/A and G/G homozygotes in young Japanese males. In contrast to this study and in agreement with Evans et al. (2000), however, we found that GG homozygotes were associated with higher BMI. Some researchers have reported no association between BMI and the UCP1-3826 G polymorphism (Evans et al., 2000; Kotani et al., 2008). Differences in study populations could strongly influence study findings. The effects of gene polymorphisms may sometimes be observed more extensively in a population under specific situations (i.e., obese population, diseased population, or ethnicity), leading to differences in data compared with earlier reports.

The GG genotype of the UCP1-3826 G polymorphism has been demonstrated to exert a significant effect on high HDL-C status and low TGs (Kotani et al., 2008). By contrast, in the present study, the GG genotype of UCP1 was related to decreased HDL-C values and high TGs, thereby suggesting higher risks of developing cardiovascular disease. The current research examined HDL-C and TGs only in obese children, while previous studies examined these parameters in healthy Japanese adults. As discussed above, obesity and ethnicity may affect findings among studies. In agreement with our study, Kiec-Wilk et al. (2002) and Oh et al. (2004) reported that GG genotype carriers present statistically significant lower HDL-C levels and higher TGs than noncarriers.

In agreement with other researchers who found no associations between UCP1-3826 A/G polymorphisms and DM2 (Brondani et al., 2014), significant differences between fasting glucose and insulin levels and HOMA-IR values according to the polymorphisms studied were not observed in this work.

In the present study, in agreement with the study of Ochoa et al. (2007), UCP2 exon 8 del/ins polymorphisms were not linked to obesity; however, homozygotes of DD in UCP2 exon 8 polymorphisms were associated with high BMI levels, inconsistent with the suggestion of no correlation between UCP2 and BMI (Evans et al., 2000). In the UCP2 polymorphism, the del allele is associated with low HDL cholesterolemia. This polymorphism has been the subject of recent reports with conflicting findings. In Spanish children, obesity was associated with a haplotype that included del alleles (Ochoa et al., 2007). Unlike in the present study, research, including a smaller sample of children, indicated that the ins variant is related to obesity and that children with the del/ins genotype have markedly higher BMIs than those with the DD genotype (Yanovski et al., 2000). The ins/ins homozygous genotype has also been correlated with high BMI levels in obese German (Evans et al., 2000) and Hungarian (Csernus et al., 2013) populations. Consistent with our findings, however, the ins allele frequency did not differ between obese and normal-weighing German children (Tu et al., 1998). Supplementary results were obtained in a study that showed carriers of the 45 bp insertion featuring higher metabolic rates than those without this insertion (Kovacs et al., 2005). This impact on energy metabolism could explain the higher obesity risk of these alleles in noncarriers than in carriers.

Although linkages have been reported between markers close to the UCP2 gene and diabetes (Fleury et al., 1997), a prediabetic phenotype in Pima Indians (Pratley et al., 1998), and insulin levels and insulin resistance (Ochoa et al., 2007), we found no association between the polymorphism and insulin levels, insulin resistance, and HOMA-IR values in obese children. These discrepancies can be partly explained by differences in study design (e.g., phenotypical measurements, characteristics of the participants, ethnic origin, age, gender, sample size, and statistical analysis).

Obese subjects who were carriers of both the GG homozygotes of the UCP1 polymorphism and the DD homozygotes of the UCP2 polymorphism were at significantly higher risk for developing obesity and low HDL cholesterolemia than those who were not. A synergistic effect of the two polymorphisms has been reported by a number of groups (Fumeron et al., 1996; Valve et al., 1998; Evans et al., 2000), in agreement with the present findings. A low HDL cholesterol level is a major risk factor for atherosclerosis (Lewis and Rader, 2005). Thus, the synergistic effect of the UCP1 and UCP2 polymorphisms may be a risk factor for cardiovascular disorders.

While UCP3-55C/T polymorphisms were not associated with obesity, CC homozygous carriers of the UCP3-55C/T polymorphism showed a risk for hypertension in obese children consistent with previous studies (Cha et al., 2007; Csernus et al., 2013), but no risk for atherosclerotic diseases, inconsistent with Meirhaeghe et al. (2000). A previous study suggested a positive correlation between UCP3-55 TT polymorphism and BMI (Otabe et al., 2000), but we did not observe this relationship because TT homozygous carriers were not found in this work. The UCP2 polymorphism has been linked to hypertension in Japanese adults (Ji et al., 2004); the polymorphism of UCP3 does not appear to show such a relationship. The 55 CT genotype of UCP3 presents increased risk of insulin resistance in obese subjects (Ochoa et al., 2007).

Conclusion

Our observations suggest that, although not the main or only factor of obesity development, the GG genotype of the UCP1-3826 A/G polymorphism could contribute to the onset of childhood obesity in Turkish children. When the synergistic effect of certain polymorphisms, each conferring a small increase in the risk to the individual, is considered, the GG genotype of UCP1 together with the DD genotype of UCP2 polymorphism may increase the risk of obesity by twofold. UCP polymorphisms may also contribute to obesity-related disorders, including dyslipidemia and hypertension. A combination of polymorphisms in one or more candidate genes may contribute to the development of most cases of obesity. Thus, obesity, similar to other metabolic disorders, is a complicated process that involves environmental, nutritional, and genetic (i.e., polymorphisms) factors. Further studies are needed to better understand the role of genetic polymorphism and functional activity of UCP genes in the pathogenesis of obesity.

In conclusion, we found that a genetic variant of UCPs appears to be associated with childhood-onset obesity and its morbidities.

Footnotes

Acknowledgment

This work was supported by Gaziosmanpasa University Scientific Research Projects Fund. Project number: 2014/53.

Author Disclosure Statement

No competing financial interests exist.

References

Boss

, Samec

, Paoloni-Giacobino

, et al. (1997) Uncoupling protein‐3: a new member of the mitochondrial carrier family with tissue‐specific expression. FEBS Lett, 408:39-42.

Brondani

, Assmann

, Duarte

, et al. (2012) The role of the uncoupling protein 1 (UCP1) on the development of obesity and type 2 diabetes mellitus. Arq Bras Endocrinol Metabol, 56:215-225.

Brondani

, Duarte

, Canani

, et al. (2014) The presence of at least three alleles of the ADRB3 Trp64Arg (C/T) and UCP1-3826A/G polymorphisms is associated with protection to overweight/obesity and with higher high-density lipoprotein cholesterol levels in Caucasian-Brazilian patients with type 2 diabetes. Metab Syndr Relat Disord, 12:16-24.

Cha

, Kim

, et al. (2007) Association of UCP2 and UCP3 gene polymorphisms with serum high-density lipoprotein cholesterol among Korean women. Metabolism, 56:806-813.

Csernus

, Pauler

, Erhardt

, et al. (2013) Uncoupling protein‐2 gene polymorphisms are associated with obesity in Hungarian children. Acta Paediatr, 102:e200-e204.

Dalgaard

, Pedersen

(2001) Uncoupling proteins: functional characteristics and role in the pathogenesis of obesity and Type II diabetes. Diabetologia, 44:946-965.

Evans

, Minouchehr

, Hagemann

, et al. (2000) Frequency of and interaction between polymorphisms in the b3-adrenergic receptor and in uncoupling proteins 1 and 2 and obesity in Germans. Int J Obes Relat Metab Disord, 24:1239-1245.

Falkner

, Daniels

(2004) Summary of the fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Hypertension, 44:387-388.

Fleury

, Neverova

, Collins

, et al. (1997) Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat Genet, 15:269-272.

10.

Fumeron

, Durack-Bown

, Betoulle

, et al. (1996) Polymorphisms of uncoupling protein (UCP) and beta 3 adrenoreceptor genes in obese people submitted to a low calorie diet. Int J Obes Relat Metab Disord, 20:1051-1054.

11.

Garlid

, Orosz

, Modrianský

, et al. (1996) On the mechanism of fatty acid-induced proton transport by mitochondrial uncoupling protein. J Biol Chem, 271:2615-2620.

12.

, Ikegami

, Fujisawa

, et al. (2004) A common polymorphism of uncoupling protein 2 gene is associated with hypertension. J Hypertens, 22:97-102.

13.

Jun

, Kim

, Lee

, et al. (2009) Effects of UCP2 and UCP3 variants on the manifestation of overweight in Korean children. Obesity, 17:355-362.

14.

Keskin

, Kurtoglu

, Kendirci

, et al. (2005) Homeostasis model assessment is more reliable than the fasting glucose/insulin ratio and quantitative insulin sensitivity check index for assessing insulin resistance among obese children and adolescents. Pediatrics, 115:e500-e503.

15.

Kiec-Wilk

, Wybranska

, Malczewska-Malec

, et al. (2002) Correlation of the −3826A >G polymorphism in the promoter of the uncoupling protein 1 gene with obesity and metabolic disorders in obese families from southern Poland. J Physiol Pharmacol, 53:477-490.

16.

Kotani

, Sakane

, Saiga

, et al. (2008) Relationship between A-3826G polymorphism in the promoter of the uncoupling protein-1 gene and high-density lipoprotein cholesterol in Japanese individuals: a cross-sectional study. Arch Med Res, 39:142-146.

17.

Kovacs

, Ma

, Hanson

, et al. (2005) Genetic variation in UCP2 (uncoupling protein-2) is associated with energy metabolism in Pima Indians. Diabetologia, 48:2292-2295.

18.

Lewis

, Rader

(2005) New insights into the regulation of HDL metabolism and reverse cholesterol transport. Circ Res, 96:1221-1232.

19.

Lobstein

, James

, Cole

(2003) Increasing levels of excess weight among children in England. Int J Obes, 27:1136-1138.

20.

Marti

, Moreno-Aliaga

, Hebebrand

, et al. (2004) Genes, lifestyles and obesity. Int J Obes, 28:S29-S36.

21.

Matsushita

, Kurabayashi

, Tomita

, et al. (2003) Effects of uncoupling protein 1 and ß3-adrenergic receptor gene polymorphisms on body size and serum lipid concentrations in Japanese women. Maturitas, 45:39-45.

22.

Meirhaeghe

, Amouyel

, Helbecque

, et al. (2000) An uncoupling protein 3 gene polymorphism associated with a lower risk of developing Type II diabetes and with atherogenic lipid profile in a French cohort. Diabetologia, 43:1424-1428.

23.

Nakano

, Shinka

, Sei

, et al. (2006) A/G heterozygote of the A-3826G polymorphism in the UCP-1 gene has higher BMI than A/A and G/G homozygote in young Japanese males. J Med Invest, 53:218-222.

24.

Neyzi

, Günöz

, Furman

, et al. (2008) Türk çocuklarında vücut ağırlığı, boy uzunluğu, baş çevresi ve vücut kitle indeksi referans değerleri [Weight, height, head circumference and body mass index references for Turkish children]. Cocuk Sagligi Hast Derg, 51:001-014.

25.

Ochoa

, Santos

, Azcona

, et al. (2007) Association between obesity and insulin resistance with UCP2-UCP3 gene variants in Spanish children and adolescents. Mol Genet Metab, 92:351-358.

26.

, Kim

, Choi

, et al. (2004) The effects of uncoupling protein-1 genotype on lipoprotein cholesterol level in Korean obese subjects. Metabolism, 53:1054-1059.

27.

Otabe

, Clement

, Dina

, et al. (2000) A genetic variation in the 5′ flanking region of the UCP3 gene is associated with body mass index in humans in interaction with physical activity. Diabetologia, 43:245-249.

28.

Pecqueur

, Cassard-Doulcier

A-M

, Raimbault

, et al. (1999) Functional organization of the human uncoupling protein-2 gene, and juxtaposition to the uncoupling protein-3 gene. Biochem Biophys Res Commun, 255:40-46.

29.

Pratley

, Thompson

, Prochazka

, et al. (1998) An autosomal genomic scan for loci linked to prediabetic phenotypes in Pima Indians. J Clin Invest, 101:1757.

30.

Rutter

(2001) Nutrient-secretion coupling in the pancreatic islet β-cell: recent advances. Mol Aspects Med, 22:247-284.

31.

Sangun Ö, Dündar

, Köşker

, et al. (2011) Prevalence of metabolic syndrome in obese children and adolescents using three different criteria and evaluation of risk factors. J Clin Res Pediatr Endocrinol, 3:70-76.

32.

Schrauwen

, Xia

, Bogardus

, et al. (1999) Skeletal muscle uncoupling protein 3 expression is a determinant of energy expenditure in Pima Indians. Diabetes, 48:146-149.

33.

Silva

, Rabelo

(1997) Regulation of the uncoupling protein gene expression. Eur J Endocrinol, 136:251-264.

34.

Sramkova

, Krejbichova

, Vcelak

, et al. (2007) The UCP1 gene polymorphism A-3826G in relation to DM2 and body composition in Czech population. Exp Clin Endocrinol Diabetes, 115:303-307.

35.

, Chen

, Winnikes

, et al. (1998) Structural organization and mutational analysis of the human uncoupling protein-2 (hUCP2) gene. Life Sci, 64:PL41-PL50.

36.

Valve

, Heikkinen

, Rissanen

, et al. (1998) Synergistic effect of polymorphisms in uncoupling protein 1 and β3-adrenergic receptor genes on basal metabolic rate in obese Finns. Diabetologia, 41:357-361.

37.

Verdi

, Tulgar Kinik

, Yilmaz Yalcin

, et al. (2015) beta-3AR W64R polymorphism and 30-minute post-challenge plasma glucose levels in obese children. J Clin Res Pediatr Endocrinol, 7:7-12.

38.

Walder

, Norman

, Hanson

, et al. (1998) Association between uncoupling protein polymorphisms (UCP2-UCP3) and energy metabolism/obesity in Pima Indians. Hum Mol Genet, 7:1431-1435.

39.

Yanovski

, Diament

, Sovik

, et al. (2000) Associations between uncoupling protein 2, body composition, and resting energy expenditure in lean and obese African American, white, and Asian children. Am J Clin Nutr, 71:1405-1420.