Heterozygote polymorphisms of ARG 16 GLY and GLN 27 GLU ADRB 2 gene is risk protective for obesity in Javanese population of Indonesia

The relationship between obesity and insulin resistance is influenced by various factors, including genetics. ADRB2 gene polymorphisms of Arg16Gly and Gln27Glu showed inconsistent results toward obesity and insulin resistance in some populations. This study aimed to investigate the frequency of the genetic polymorphisms of Arg16Gly and Gln27Glu of the ADRB2 gene and to analyze risk factors for obesity and insulin resistance in Javanese sample population. This study was conducted on 100 control and 100 obese subjects. Subjects were measured for body weight and height to determine body mass index. Waist hip ratio was calculated by dividing the size of waist circumference by hip circumference. Fasting blood glucose was measured with GOD-PAP methods, and plasma insulin levels was measured using ELISA. Arg16Gly and Gln27Glu ADRB2 gene polymorphisms were detected using the PCR-RFLP method. Results of polymorphisms of Arg16Gly showed that AG genotype decreased the risk of obesity with OR 0.264 (CI 95% : 0.119 – 0.585), p = 0,001 compared to the AA genotype. CG genotype of Gln27Glu increased the risk of obesity with RR 2.082 (CI 95% : 1.786 – 2.427). CC and CG genotypes toward GG genotype in the obese group had significant differences for plasma insulin, but were not significantly different for BMI, waist hip ratio, fasting blood glucose, and HOMA-IR. The combination of Arg16Gly and Gln27Glu AG+CG decreased the risk of obesity compared to AA+GG; however, it was not associated with insulin resistance.


INTRODUCTION
Obesity is caused by an imbalance between food intake and energy use.This imbalance is due to the complex interactions between dietary habits, lack of physical exercise and genetic background (Marti et al., 2004).Data from the Indonesia Basic Health Research (Riskesdas) in 2013 showed that the prevalence of obesity in the adult age group is 15.4% (Kemenkes, 2003).Insulin resistance is a decrease in insulin activity in skeletal, liver, and adipose tissue (Saltiel and Kahn, 2001).Insulin and catecholamines are the major hormonal regulators of fat cell metabolism in humans.The function of hormones in fat cells changes in obese people.Insulin stimulates glucose uptake, lipid synthesis of glucose (lipogenesis) and inhibits the hydrolysis of triglycerides into glycerol and free fatty acids (Large and Arner, 1998).Catecholamines stimulate lipolysis through adrenoreceptors-β1, β2, and β3, but inhibit the effects of lipolysis mediated by α2-adrenoreceptors (LaFontan and Berlan, 1995).
The β2-adrenergic receptor gene (ADRB2) is expressed in many tissues including the liver, adipose tissue, smooth muscle in the bronchus and intestine.The ADRB2 gene regulates energy use by stimulating lipolysis and thermogenesis through activation of catecholamine induction from adenylate cyclase through the action of G protein (Barbe et al., 1996;Naka et al., 2013).Common ADRB2 polymorphisms are changes of glutamine into glutamic acid at codon 27 (Gln27Glu) and arginine to glycine at codon 16 (Arg16Gly).Some studies indicated that Arg16Gly and Gln27Glu have been shown to be associated with both obesity and body mass index (BMI) (Gjesing et al., 2007), but other studies found no association between Arg16Gly and Gln27Glu with obesity or BMI (Pereira et al., 2003;Bengtsson et al., 2001).
The Gln27Glu polymorphism has a significant association with increased waist circumference and free fatty acids (Park et al., 2008).One of meta-analysis for Arg16Gly and Gln27Glu showed that the 27 Glu allele was a significant risk factor for obesity in Asia Pacific and American populations (Jalba et al., 2008).
Based on this background, the aim of this study was to determine the difference of genotype frequency encoding ADRB2 gene on codon 16 Arg16Gly and codon 27 Gln27Glu between obese and control groups, and to study Arg16Gly and Gln27Glu ADRB2 gene polymorphism as a risk factor for obesity and insulin resistance.

Subjects
The subjects included 200 healthy volunteers: 100 obese (50 males and 50 females) and 100 control (50 males and 50 females) with age 18 -35 years' old, Javanese people.Subjects were categorized as control group with BMI 18.5 -22.9 kg/m 2 and obesity group with BMI >25 kg/m 2 .These subjects were recruited in the Universitas Gadjah Mada and University of Muhamadiyah Purwokerto.The study was approved by the Medical and Health Research Ethics Committee, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada with reference number: KE/FK /0644/EC/2017.Subjects were excluded if they were not Javanese people, under 18 years and over 35 years, had diseases associated with metabolic syndrome, weight loss program, smoking, and had fasting blood glucose ≥ 125 mg/dL.

Anthropometric measurements
Height and weight of subjects were first measured and their BMI scores calculated.Waist hip ratio was calculated by dividing the size of waist circumference by hip circumference (Chan et al., 2003).

Measurements of blood glucose, plasma insulin levels and HOMA-IR
Five milliliters fasting blood was collected from study subjects and inserted in the EDTA test tube.Blood was centrifuged to get plasma and buffy coat.Plasma is used to measure the blood glucose and insulin levels.Buffy coat was taken for DNA isolation and genotyping.Measurements of blood glucose used GOD-PAP enzymatic colorimetric tests.Measurements of plasma insulin concentrations were based on ELISA KIT Insulin DRG® protocol.The HOMA-IR was calculated by multiplying plasma Insulin (µIU/mL) x fasting blood glucose (mmol/L) and dividing by 22.5 (Yamada et al., 2012).

Genotyping
Genomic DNA was prepared from buffy coat using PROMEGA kit.For single nucleotide polymorphisms (SNPs) in the Arg16Gly and Gln27Glu ADRB2 gene, the primers were as follows: (forward primer 5' CAGCGCCTTCTTGCTGGCACCCCAT-3' and reverse primer 5'-CTGCCAGGCCCATGACCAGATCAG-3'.Polymerase chain restriction (PCR) process began with an initial denaturation of 94°C for 2 min, followed by 35 cycles with temperature of 94°C for 30 s, annealing 65°C for 45 s, elongation at 72°C for 1 min, and final elongation at 72°C for 7 min.The resulting PCR product was 242 bp (Aynacioglu et al., 1999).The PCR product was then cut by restriction enzyme and incubated for 16 h at 37°C.The Eco130I restriction enzyme was used for detection of the Arg16Gly polymorphism and Fnu4HI restriction enzyme for detection of the Gln27Glu polymorphism.Digestion products were for Arg16 = 242 bp, Gly16 = 214 bp, Gln27 = 181 bp, and Glu27 = 236 bp.

Statistical analysis
Characteristics of subjects were first tested for normality with the Kolmogorov-Smirnov test.The comparison of BMI, waist hip ratio, fasting blood glucose, insulin level, and HOMA IR mean between obese and control groups used independent sample t-tests when the data were normally distributed.Differences in allele and genotypes frequencies in each obese and control group were analyzed with Chi-square tests.Risk factors for each genotype to BMI and HOMA-IR between obese and control groups were tested with odds ratios.The comparison of BMI, waist hip ratio, fasting blood glucose, plasma insulin levels and HOMA IR mean between groups of genotypes was performed with independent sample ttest.

RESULTS
This study was conducted with 200 people who had met the inclusion criteria and signed an informed consent form.Characteristics of obese and control group were presented in Table 1.The subjects were divided into two groups: control group (non obese) with body mass index  The genotyping of ADRB2 Arg16Gly gene was performed by PCR-RFLP method.The resulting PCR product was 242 bp; thereafter, it was cut with the Eco130I enzyme.Genotypes found in ADRB2 Arg16Gly gene polymorphism in this study were AA, AG, and GG.The result of analysis of ADRB2 Arg16Gly genotype polymorphism can be seen in Figure 1, whereas the genotypes found in ADRB2 Gln27Glu gene polymorphism in this study were CC, CG, and GG.GG genotype in this research were wild type because they appeared more than the CC genotype with 236 bp fragment length, CG genotype with 236 bp and 181 bp fragments, and CC with 181 bp fragment length.The results of genotype analysis of ADRB2 Gln27Glu polymorphism can be seen in Figure 2.
Based on the data in Table 2, the genotype frequency distribution of obese was AA (Arg16Arg): 27%; AG (Arg16Gly): 62%; and GG (Gly16Gly): 11%.The genotype distribution of the control group was AA (10%), AG (87%) and GG (3%).Genotype AG can reduce the risk of obesity by 0.264 (CI: 95% = 0.119 -0.585) with a value of p = 0.001 compared to genotype AA.The genotype distribution of the obese group was GG (Glu27Glu): 85%;  CG (Gln27Glu): 11%; CC (Gln27Gln): 4%, whereas the genotype distribution of control group was GG (92%) and CC (8%).The genotype frequency of CG has an obesity risk of 2.082 (CI: 95% = 1.786 -2.427) compared to GG genotype.The distribution of genotypes observed was compared with expected results by Hardy-Weinberg calculations.Based on χ2 test, it was found that the result of observation and expected result shows that the result was not significant.Hardy-Weinberg equilibrium test results on the genotype of ADRB2 gene polymorphism Arg16Gly and Gln27Glu in Javanese population showed no significant differences between the genotype distributions of the research result (Table 2) with the population of HWE (Table 3).
The role of ADRB2 Arg16Gly gene polymorphism on BMI, waist hip ratio, fasting blood glucose, plasma insulin, and HOMA-IR in control and obesity group were tested using independent sample t-test.The genotype AA with genotype GG and AG did not have significant differences in body mass index, waist hip ratio, fasting blood glucose, plasma insulin and HOMA-IR in the control group and obese group.The CC and CG genotypes toward GG genotype in the obese group had significant differences in plasma insulin (p : 0.011), however, it was not significantly different for BMI, hip ratio, fasting blood glucose levels and HOMA-IR.The CC + CG genotype has a tendency towards insulin resistance with a higher mean HOMA-IR value than the GG genotype although not different.
Based on Table 5, the results of the analysis combination of AG+CG toward AA+GG showed OR value of 0.280 (CI 95%: 0.119 -0.659) with p = 0.002.GG+GG haplotype toward AA+GG haplotype showed the OR value 1.929 (CI 95%: 0.345 -10.767) with p value = 0.371 was not significantly different between obese and control groups.

DISCUSSION
The combination analysis of ADRB2 gene genotypes Arg16Gly and Gln27Glu aimed to know the difference in frequency between the obese and control groups and their relation to obesity risk.The combination of genotypes in Arg16Gly and Gln27Glu (AG +CG) ADRB2 gene polymorphisms toward AA + GG was significantly different between obese and control groups with OR 0.280 (CI: 95% : 0.119 -0.659), p = 0.002.The combination of AG + GG had a protective risk on obesity compared to the AA + GG haplotype.This study was shows Haplotype AG + CG had a protective risk to obesity.Haplotypes that provide protective risks are associated with catecholamine reduction by inducing cAMP production in recombinant cells and decreasing insulin sensitivity in human fat cells (Drysdale et al., 2000;Eriksson et al., 2004).The mismatch between single nucleotide polymorphism (SNP) analysis and haplotype analysis of the ADRB2 gene against obesity provides a debatable outcome (Rosmond, 2003).
Protective haplotypes and other haplotypes had no effect on cAMP production on mononuclear cells in peripheral blood (Lipworth et al., 2002).Human fat cells have four ADRB subtypes that regulate lipolysis (Large and Arner, 1998).The mechanism is related to the function of other adrenoceptors in adipocytes.Another thing is the existence of gene and environment interactions such as fitness and diet (Jiao et al., 2005).ADRB2 haplotype may have a direct influence on glucose metabolism.For example, ADRB2 triggers blood flow (Barbe et al., 1996).Differences in blood flow between haplotypes ADRB2 will affect the delivery of glucose and insulin in the peripheral tissues, thus affecting glucose tolerance.The β-Adrenergic stimulus also produces insulin resistance in skeletal and fatty muscles (Prior et al., 2011).This causes a different stimulus of the haplotype ADRB2 to explain the independent relationship of the haplotype ADRB2 with insulin resistance.The ADRB2 genotype may affect receptor function, receptor density, or efficiency.Reduced expression of the receptor may result in inefficient lipolysis of adipose tissue and the accumulation of excess fat over time (McGraw et al., 1998).The combination of Arg16Gly and Gln27Glu did not appear to affect being overweight and obese, or subjects with BMI > 30 kg/m 2 when compared with subjects with BMI <25 kg/m 2 (Gjesing et al., 2007).The ADRB2 haplotype study conducted by Lima et al. (2007) in both white and black male and female subjects was not associated with BMI.
The obese group had a higher mean than the control group (Table 1).Body mass index was associated with high fasting glucose levels in men and elevated concentrations of insulin in women (Patel and Abate, 2013).Fasting blood glucose levels and plasma insulin levels were elevated in obese subjects (Nguyen and Nguyen, 2008).When BMI increases, HOMA-IR will also increase (Martinez et al., 2017).The data in this study support the previous studies where if the obesity group had a higher mean of BMI, then the value of HOMA IR will also be high when compared with the control group.The combination of ADRB2 genotype to insulin resistance calculated by HOMA-IR in this study did not differ high insulin resistance.In several previous studies, ADRB gene polymorphisms were associated with serum insulin levels and insulin resistance but were not associated with obesity in Swedish women (Mottagui et al., 2008).The HOMA IR in the CC+CG genotypes of the obese group had a mean value (3.49±2.84µIU/mL) greater than GG genotype (2.51±1.90µIU/mL).Association between the combination of polymorphism and metabolic characteristics requires a larger number of samples (Tsunekawa et al., 2011).ADRB2 gene polymorphism is often found in obese individuals (Large et al., 1997), although the link between polymorphism with body weight, glucose tolerance, insulin sensitivity has shown mixed results (Oberklofer et al., 2000;Prior et al., 2011;Echwald et al., 1998).
In this study, CC and CG genotypes had a higher mean BMI when compared to the GG genotype (Table 4).Research conducted by Kawamura et al. (2001) showed no association between the ADRB polymorphism of Gln27Glu gene against obesity.The results of research conducted by Oberklofer et al. (2000) also showed the polymorphism of Gln27Glu gene ADRB2 not related to obesity in Austrian women.Similar results in a study conducted by Gjesing et al. (2007) suggests that the OR allele Glu27 is 0.92 and is not associated with obesity.Frequency of the CG (Gln27Glu) genotype in this study increased obesity risk of 2.082 compared to GG genotype.These results indicated that the subjects with Gln27Glu variation were at risk of becoming obese by 2.082 times compared to the GG (wild type) genotype.The genotype frequency of CG was most found in case group of Austria (47.5%) and in Spain with (47.3%).The genotype frequency of CG was found higher in the control group in Sweden (53.4%) and Denmark (46.5%) (Zhang et al., 2014).The GG genotypes in Sweden and Denmark were found higher in the case group with significant comparisons in Sweden (24.4%) vs (3.4%), and Denmark (30.2%) vs (16.5%), whereas in Austria this study showed the genotype frequency of GG most commonly found in the control group.Research Kim et al. (2002) did not find GG genotypes in both case and control groups.There was no association between the polymorphism of Gln27Glu of ADRB gene and obesity (Kawamura et al., 2001).The obese phenotype in humans appears to be determined not only by a number of different genes but also because of links to other diseases, genetic variations, and environmental factors (Gjesing et al., 2007).The Glu27 allele has a tendency to increase BMI (Large et al., 1997), waist-to-hip ratio (Hellstrom et al., 1999), type 2 diabetes mellitus (Ishiyama-Shigemoto et al., 1999), and can suppress lipid oxidation (Macho-Azcarate et al., 2003).Echwald et al. (1998) found that ADRB2 Glu27 polymorphisms were not associated with obesity in the Ducati Caucasian male population.The Glu27 allele group had low lipolysis as measured by plasma glycerol.Echwald (1998) and Oberkofler et al., (2000) found no association between ADRB2 and obesity.
The AG (Arg16Gly) genotype decreased the risk of obesity by 0.264 compared to the AA (Arg16Arg) genotype.Arg16Gly ADRB2 gene polymorphism in some populations in the world showed different results.The

Table 1 .
Characteristics of obese and control groups.
n, Number of subjects.Data were presented in mean ± SD. *Independent sample t test, p<0.05 was significantly different.

Table 2 .
Codon 16 and 27 polymorphisms of ADRB2 gene in obese and control subjects.

Table 3 .
Observed and expected genotype distribution of ADRB2 in obese and control group.
Chi square test; p < 0.05 : was significantly different.

Table 5 .
Haplotype of Arg16Gly and Gln27Glu polymorphisms in ADRB2 gene in obese and control subjects.