The Study On The Relationship Between G1057D Variants Of IRS-2 Gene And T2DM

ObjectiveTo investigate whether the G1057D variants of insulin receptor substrate -2 (IRS -2) gene have relation with type 2 diabetes mellitus (T2DM) , the role of obesity especially central obesity in T2DM and whether the healthy subjects with G1057D variants of IRS - 2 have the potential genetic trend of developing into T2DM. Furthermore to establish the academic basis of early finding, correct classification, effective therapy, reasonable prophylaxis and exploitation of new drugs for T2DM.MethodsThe Han population in Liaoning area was selected as the study subjects. Three studies have been performed according to the objective of study.The first study:1. Subjects; there were 221 healthy subjects aged 40 -60 years old with normal oral glucose tolerance test (OGTT) and no family history of DM. Waistline, hip, body weight and height were measured, waist/hip ratio (WHR) and body mass index (BMI) were calculated. According to BMI, the 221 subjects were divided into 2 groups; obese group with BMI≥25kg/m~2(excluding the o-besity and obesity syndrome related to inherit, endocrinology and central nervous system) and non -obese group with BMI≤25kg/m~2. There were 113 subjects in obese group and 108 subjects in non - obese group. The two groups were matched in sex and age.2. Methods: G1057D variants of IRS -2 were detected by PCR - RFLP a-nalysis. And at the same time, some indexes of insulin secretion and insulin action associated with the mechanism of T2DM were analyzed, including fasting plasma glucose (FPG) , fasting insulin (FINS), fasting C peptide (FCP) , 2 hour plasma glucose (2hPG) , 2 hour insulin (2hINS) , 2 hour C peptide (2hCP) , homeostasis model assessment - {3 cell function ( HOMAB) and home-ostasis model assessment - insulin resistance ( HOMAIR). The correlation of G1057D variants of IRS -2 with BMI, WHR, and blood pressure was also investigated.(1) the detection of variants of IRS - 2 gene: (I) preparation of genomic DNA: 5ml of fasting venous blood was obtained in test tube containing EDTA. White blood cells were isolated to extract genomic DNA by traditional saturated sodium chloride method. @Polymerase chain reaction (PCR) amplification: PCR primer was obtained by on - line software primer 3 and synthesized by Shanghai Shenggong biology and project company. Upstream primer was5' - gtccccgtcgtcgtctct - 3' and downstream primer was 5' - ctcgactcccgacacctg -3' The reaction system contained template DNA, buffer, dNTP, primer, and Tag enzyme. PCR included an initial denaturation at 94'C for 5 minutes, followed by 30 cycles of denaturation at 94X. for 50 seconds, annealing at 581 for 50 seconds, extension at 72*^0 for 7 minutes. PCR amplification was performed by a PCR thermocyler (USA Perkin - Elmer type 9600). (3)RFLP detection; PCR products were identified by 1.2% agarose gel electrophoresis and then digested with restriction enzyme Hae E (NEB) at 37X1 overnight. The digested products were stained with ethidium bromide after 8% polyacrylamide gel electrophoresis, and detected with ultraviolet lamp. The images of electrophoresis were scaned by computer.(2) biochemical assays and radio - immune analysis: ? PG ( oxidase method) ; d) insulin and c —peptide detection (radio - immunity analysis, reagents and chemiluminescence immunoassay equipment were supplied by USA -SP company. ) ? blood triglyceride and cholesterol levels (enzyme method), HDL - C and LDL - C levels were determined by dextran sulphate - manganium precipitation method and calculated by formula. @ the secretion and action ofinsulin was evaluated by homeostasis model assessment - (3 cell function ( HOM-AB) and homeostasis model assessment - insulin resistance (HOMA - IR). HOMAB =20 x FINS/FPG -3.5, HOMA - IR = FINS x FPG/22.5. (HOMAB and HOMA - IR should be natural logarithm - transformed).(3) Statistical analysis; The frequency of gene variants (genotype and al-lelic frequency) were calculated by gene counting method, the colony representation of sample was confirmed by Hardy - Weinberg balance analysis. All normal distribution measurement data were expressed as means ± SD. Skewed distribution data were analyzed after logarithm - transformed. Student's t test, one - way analysis of variance, and x tests were appropriately performed for comparison of groups. The relationships were analyzed by Logistic regression. All data were analyzed by SPSS version 10.0, p values less than 0.05 were considered significant.The second studyTwo hundred and thirty five subjects were selected randomly, among which 116 were no related type 2 diabetic patients including 60 men and 56 women. Diabetes mellitus was diagnosed and classified according to American Diabetes Association criterion in 1997. All the patients recruited from the first affiliated hospital of China Medical University. 119 healthy controls including 58 men and 61 women were subjects with normal OGTT results and no family history of type 2 diabetes. The controls were matched with the cases in age, sex, and BMI. All the analytic methods were the same as those in the first study except for non -conditional Logistic regression that was performed for correlation analysis.The third studyTwo hundred and twenty five obese subjects were selected randomly, among which there were 112 no related type 2 diabetes who came from the first affiliated hospital of China Medical University and 113 healthy controls. The diagnostic criteria of diabetes mellitus, obesity and experimental methods were the same as those in the first and the second studies.ResultsThe results of the first study:(1) polyacrylamide gel electrophoresis results of the fragments of PCR -amplified IRS -2 G1057D obtained using restriction enzyme Haell: normal ho-mozygote GG genotype had two bands; 183bp and 102bp; heterozygote GD genotype had three bands: 285bp, 183bp and 102bp; mutational homozygote DD genotype had one 285 bp band.(2 ) The difference between observed number and expected number of different genotypes of IRS - 2 gene G1057D site had no significance by the goodness of fit test of Hardy - Weinberg law, which suggested that the distribution of allele frequency fits Hardy - Weinberg equilibrium. The frequency of G1057D variant was 29. 6% in all subjects with 24. 3% in obese group and 35. 2% in non - obese group (x = 6.233,P=0.013). The difference of the genotypes of IRS -2G1057D between the obese group and non -obese group was significant by x2 test (x2 =6. 075, P = 0.048). The frequency of DD genotype was significantly lower in obese group than that in non -obese group 6. 2% vs 14. 8% (x2 = 4.401, P =0.040).(3) In the non - obese group, FPG and 2hCP of DD genotypes were lower than those of GG genotypes (P <0. 05, P <0. 01) as well as 2hPG and 2hINS also had the descending trends ( P > 0. 05 ) compared with DD genotypes ( P > 0. 05 ). However HOMAB of DD genotypes was higher than that of GG genotype (P <0.01). (4) In the obese group, WHR, HOMAIR, and 2hCP were higher than those of GG genotypes ( P <0.05, P <0. 01, P <0. 01) , and the ratio of central obesity was 5/7.The results of the second study:(1) The results of electrophoresis were the same as that of the first study.(2) The frequency of G1057D variant was 29.8% in general subjects with 23. 3% in type 2 diabetic group and 36. 1% in normal control group (x2 = 9.288,P =0.002). The difference of the genotypes of IRS -2G1057D between the two groups was significant. The frequency of genotype of IRS - 2G1057D al-leles was significant difference between diabetic and normal groups (x -^JOj^ 195, P =0.006). The frequency of DD genotype was significantly lower in DM group than that in control group with 4.3% vs 16% (x2 =8.70, P=0.004).(3) The Logistic regression analysis showed that the OR values for GD and DD genotype of IRS -2 were 0. 832 (95% confidence interval: 0.478 -1.446, P =0.514) and 0.244 (95% confidence interval: 0.084-0.703, P =0.009) respectively which demonstrated that the risk of T2DM decreased with the increasing of D1057 allele and DD genotype was the protective factor of T2DM.(4) In the control group, FPG of DD genotype was lower than that of GG genotype (P <0.05). HOMAB of GD and DD genotypes were higher than that of GG genotype (P <0.05,P <0.01). (5) In the DM group, HOMAIR of DD genotype had an ascending trend and HOMAB of DD genotype had a descending trend compared with that of GG genotype ( P > 0.05 ). But BMI especially WHR of DD genotype were significantly higher than those of GG genotype (P <0.05, P <0. 01). And in the 5 cases with DD genotype, there were 4 cases with central obesity and 1 female case with WHR =0. 86.The results of the third study .-(1) The results of electrophoresis were the same as that of the first study.(2) The difference between observed number and expected number of different genotypes of IRS -2 gene G1057D site had no significance by the goodness of fit test of Hardy - Weinberg law, which suggested that the distribution of allele frequency fits Hardy - Weinberg equilibrium. The frequency of G1057D variant was 28.7% in all the obese subjects with 33.5% in obese T2DM group and 23. 9% in obese control group (x2 = 5.058 ,P =0.025). The different genotypes of IRS - 2G1057D site of obese T2DM group and obese control group were analyzed by x2 test ( x2 = 5. 40, P = 0. 067). The frequency of DD genotype in obese T2DM group was significantly higher than that in obese control group withl3.4% and 5.3% respectively (x2 =4.344, P= 0.041).(3) The Logistic regression analysis showed that the OR values for GD and DD genotype of IRS -2 were 1.246(95% CI: 0.710 -2.185, P =0.443) and 3.991(95% CI: 1.366-11.366, P=0.011) respectively which indicated that the risk of obese T2DM increased with the increasing of D1057 allele and DDgenotype was the risk factor of obese 12DM.(4) WHR of GD and DD genotypes in obooe diabetio group presented an fecending trend compared with that of oboao control group (P >0.05)-.(4) In the obese DM group, WHR and HOMAIR of DD genotype were higher than those of GG genotype ( P < 0.01) , but HOMAB were lower than that of GG genotype ( P < 0. 005). Also 2hPG ,2hINS and 2hCP of DD genotype were all higher than those of GG genotype ( P < 0.05 ). The ratio of central obesity of DD genotype was 12/15.(5) In obese control group, WHR, 2hINS, and HOMAIR were higher than those of GG genotype, and HOMAB was lower than that of GG genotype ( P<0.05).ConclusionDD genotype of G1057D variants of IRS -2 shows double qualities. The insulin sensitivity of non - obese subjects with DD genotype increases, which indicates that DD genotype may be the protective factor of T2DM. The insulin sensitivity of obese especially central obese subjects with DD genotype decreases, which indicates that DD genotype may be the risk factor of obese T2DM. All of these demonstrated that the relationship between G1057D and type 2 diabetes are mediated by obesity. Therefore the high - risk population who are obese especially central obese subjects with DD genotype should be closely monitoredand follow - up.