AIM: To confirm the hypothesis that polymorphisms of the uncoupling protein 3 (UCP3) gene are associated with the occurrence of nonalcoholic fatty liver disease (NAFLD).

METHODS: A total of 250 NAFLD patients (147 males and 103 females) and 200 healthy individuals who served as controls (control, 109 males and 91 females), aged between 6 and 16 years were enrolled in this study. The four non-synonymous single nucleotide polymorphisms (SNPs) in the UCP3 gene polymorphisms of rs1726745, rs3781907, rs11235972 and rs1800849, were genotyped using MassArray. Body mass index (BMI), waist and hip circumference, blood pressure (BP), fasting blood glucose (FBG), insulin and lipid profiles were measured and B-ultrasound examination was performed in all subjects.

RESULTS: NAFLD patients showed risk factors for metabolic syndrome: elevated BMI, waist-to-hip ratio, BP, FBG, homeostasis model assessment-estimated insulin resistance, total triglyceride, total cholesterol and low-density lipoprotein-cholesterol, while decreased high-density lipoprotein-cholesterol level compared with the control group. The GG genotype distributions of rs11235972 in the NAFLD group differed significantly from that in the control group. We found that waist circumference between CC (58.76 ± 6.45 cm) and CT+TT (57.00 ± 5.59 cm), and hip circumference between CC (71.28 ± 7.84 cm) and CT+TT genotypes (69.06 ± 7.75 cm) were significantly different with and without rs1800849 variation (P < 0.05).

CONCLUSION: A higher prevalence of rs11235972 GG genotype was observed in the NAFLD group compared with the control group. No differences were observed for the other SNPs. However, there was a significant difference in body height in addition to waist and hip circumference between the CC (mutant type group) and CT+TT group with and without rs1800849 variation.

Nonalcoholic fatty liver disease (NAFLD) is a clinicopathologic condition characterized by abnormal lipid deposition in hepatocytes in the absence of excess alcohol intake. NAFLD comprises a wide spectrum of liver damage, including simple steatosis, steatohepatitis, fibrosis or even cirrhosis of the liver[1]. NAFLD does not only impact adults, but is also one of the major causes of liver diseases in children[2]. There are few population-based prevalence studies of pediatric NAFLD. Some studies have suggested a prevalence of 2.6%-9.6% for suspected NAFLD among children and adolescents in the United States[3,4] and Asia[5,6]. NAFLD has been shown to be associated with metabolic syndrome (MetS), which comprises obesity, type 2 diabetes, dyslipidemia and high blood pressure (BP) with insulin resistance being the central mechanism[7,8]. Theoretically, many variations in candidate genes related to MetS may contribute to the pathogenesis of NAFLD, such as genes related to insulin resistance and genes influencing hepatic free fatty acid metabolism. Elucidation of genetic factors that predispose an individual to NAFLD may lead to the development of non-invasive biomarkers for the early diagnosis of NAFLD and may allow early preventive and therapeutic strategies for those at the high risk.

Uncoupling protein 3 (UCP3) gene is located on chromosome 11q13. UCP3 is a mitochondrial anion carrier protein with a highly selective expression in skeletal muscle, a major site of thermogenesis in humans, which makes an attractive target for studies into the regulation of body weight. Reduced expression of UCP3 decreases energy expenditure and increased expression of UCP3 mRNA in muscle is related to an increase in the metabolic rate and to a lower body mass index (BMI)[9,10]. Therefore, UCP3 may be involved in obesity, given the role of UCP3 in the regulation of energy and lipid metabolism. Genetic variants of UCP3 have been identified, and specifically polymorphisms of 55C/T may impact type 2 diabetes mellitus (T2DM), obesity and weight gain[11-13]. This study confirmed the hypothesis that polymorphisms of the UCP3 are associated with the occurrence of NAFLD. These variations could be useful for the diagnosis and/or prognosis of NAFLD, although the functional significance of UCP3 polymorphisms is not clear.

The clinical features of the NAFLD and control groups are shown in Table 1. There was no significant difference in age and gender (P > 0.05). NAFLD patients showed most of the risk factors for the MetS: elevated BMI, waist-to-height ratio (WHR), BP, FBG, HOMA-IR, TG, TCHO and LDL, while decreased HDL level compared to control group.

The distributions of the four SNPs (rs1726745, rs3781907, rs11235972 and rs1800849) obeyed the Hardy-Weinberg equilibrium in all subjects. The genotypic distributions of the four loci in the UCP3 gene are shown in Figure 1 and Table 2. The genotype distributions of rs11235972 in the NAFLD group differed significantly from that in the control group.

When all variables were put into multivariate logistic regression analysis, higher BMI and HOMA-IR were risk factors for the development of NAFLD (Table 3). The clinical features in subjects with or without rs1800849 variation are shown in Table 4. There were no significant differences in anthropometric and biomedical variables with or without rs11235972 variation in the NAFLD and in the control groups, respectively. We found body height between the CC (142.93 ± 13.08 cm) and CT+TT (139.38 ± 12.10 cm), waist circumference between the CC (58.76 ± 6.45 cm) and CT+TT (57.00 ± 5.59 cm) and hip circumference between the CC (71.28 ± 7.84 cm) and CT+TT (69.06 ± 7.75 cm) genotypes were significantly different in the control group with and without rs1800849 variation (P < 0.05).

When all genotypes of the four loci were entered into haplotype analysis using Haploview, only two haplotypes’ alleles i.e., GC and AT frequencies were accepted for assessment between the NAFLD and control groups. However, this did not reach the significance as an independent risk factor for NAFLD (Table 4 and Figure 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Linkage Disequilibrium mapping around the uncoupling protein 3 between nonalcoholic fatty liver disease and control groups.

Previous reports have demonstrated that the prevalence of NAFLD increased 10%-80% in individuals with obesity, 35%-90% in individuals with T2DM, 30%-56% in individuals with hypertension, and 26%-58% in individuals with dyslipidemia[16-18]. The prevalence of the MetS among subjects with NAFLD is 17%-36%, depending on gender and criteria used[19]. In the present study, we found that an increased risk of NAFLD was significantly associated with BMI, WHR, BP, FBG, HOMA-IR, TG, TCHO and LDL, and decreased HDL level. In agreement with these studies, Iacobellis et al[20] reported that a BMI evaluation may be useful in identifying those children at higher risk for disease progression. In the present study, children with NAFLD had an elevated WHR, a surrogate marker for visceral fat. Visceral fat is closely correlated with hepatic TG content, elevated ALT, liver inflammation, and fibrosis[21-23]. Central obesity is a better measure of predisposition to insulin resistance, and is more closely associated with NAFLD.

UCP3 is considered to be associated with obesity, given the role for UCP3 in the regulation of energy and lipid metabolism. However, the effect of genetic polymorphisms in UCP3 on the pathogenesis of NAFLD has not been clearly documented. In our study, the GG genotype distribution of rs11235972 in the NAFLD group differed significantly from the control group. The results from the literature are controversial. Yoon et al[24] genotyped 6 polymorphisms of UCP3 among overweight female subjects (n = 458), and genetic effects on BMI and changes after a very low calorie diet were examined. They found that several polymorphisms in the UCP2-3 gene cluster showed associations with changes in BMI and fat mass. Hamada et al[25] determined whether the UCP3-55 C/T SNP was associated with obesity according to the criteria for Japanese (BMI ≥ 25 kg/m²) and serum HDL-C levels in the general population. Subjects with the T/T genotype had significantly higher HDL-C levels than those with the C/C genotype or C/T genotype. Furthermore, subjects with the T/T genotype had a significantly lower BMI than those with the C/C genotype[25]. Salopuro et al[26] found that the UCP3 gene variant rs3781907 was associated with increased serum TCHO and LDL cholesterol levels. The rs1726745, rs11235972 and rs1800849 variants in the UCP3 gene are associated with serum total and LDL-cholesterol at baseline. However, de Luis et al[27] did not demonstrate an association between the -55CT polymorphism of the UCP3 gene and fat distribution in obese patients. There might be true variability in the association among different populations, particularly different ethnic groups.

A common promoter polymorphism has also been identified in the UCP3 gene (rs1800849), a rare allele associated with obesity in a recessive manner in several studies[10-12]. Moreover, the rs1800849 allele is associated with a higher WHR[28], but no association between rs1800849 and WHR existed in the current study. We showed a significant difference in height, and waist and hip circumference between the CC (mutant type group) and CT+TT group with and without rs1800849 variation (Table 5).

One limitations in this study was that we used abdominal ultrasonography to diagnose NAFLD, although validation ultrasonography has a sensitivity of 91.7% and a specificity of 100%[29]. The diagnosis of NAFLD was based on ultrasound and was not confirmed by liver biopsy due to the invasive procedure usually not initially performed and ethical considerations. Thus, surrogate markers are commonly used, such as transaminases and imaging techniques. Computed tomography is more specific but is not used for screening of fatty liver in obese children. Magnetic tomography is more useful in adults and not appropriate for children due to ionizing radiation. Magnetic resonance imaging and 1H-MRS have the greatest accuracy to determine hepatic fat content, but are rarely used due to high costs. A higher prevalence of the rs11235972 GG genotype was noted in the NAFLD group compared with the control group; no differences were observed for the other SNPs. BMI and HOMA-IR increased the risk of NAFLD. Moreover, no increased risk for developing NAFLD was found to be associated with the rs1800849 variant based on multivariate analysis. A significant difference in height, and waist and hip circumference between the CC and CT+TT group with and without rs1800849 variation was demonstrated.

We sincerely thank the parents and children for participating in this study. We thank the nursing staff of our department for their dedicated care of these young patients during the collection and evaluation of blood samples. The authors thank Mr David Cushley for carefully reviewing the manuscript.