GENETIC TESTING AND MOLECULAR BIOMARKERS Volume 20, Number 3, 2016 ª Mary Ann Liebert, Inc. Pp. 130–136 DOI: 10.1089/gtmb.2015.0250
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Association of Apolipoprotein A5 Gene Polymorphisms with Metabolic Syndrome in the Korean Population Young Ree Kim1 and Seung-Ho Hong2
Aims: Functional defects of the ApoA5 protein have been identified as risk factors for hypertriglyceridemia, vascular diseases and susceptibility to metabolic syndrome (MetS). These associations are neither strong nor consistent in all populations studied. In this study, we investigated the association between the ApoA5 -1131T>C and -12,238T>C polymorphic loci in Korean patients with MetS. Methods: A total of 1074 subjects, including 415 patients with MetS and 659 healthy control subjects, were enrolled to investigate the affect of ApoA5 polymorphisms on risk of MetS. Genotyping of the ApoA5 polymorphisms was performed by polymerase chain reaction–restriction fragment length polymorphism techniques. Results: The CC genotype and the dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models of the -1131T>C polymorphism were associated with increased MetS susceptibility ( p < 0.001, p = 0.018, and p = 0.002, respectively). The association was malespecific when stratified by gender. With regard to the -12,238T>C polymorphism, the TC and CC genotypes and the dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models were frequently found in the patient group, compared with the control group ( p = 0.001, p < 0.001, p < 0.001, and p = 0.031, respectively). The T-C, C-T, and C-C haplotypes of the ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with an increased risk for MetS ( p < 0.001, p = 0.001, and p < 0.001, respectively). The variant of the ApoA5 -1131T>C polymorphism was also associated with increased triglyceride (TG) levels. Dominant models of ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with the risk components of MetS by the stratification analysis. Conclusion: The -1131C and -12,238C variants and the C-containing haplotypes of ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with higher risk for MetS in the Korean population. The -1131C variant was also associated with the increased level of TG. Introduction
M
etabolic syndrome (MetS) is developed due to abnormality of waist circumference (WC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), blood pressure, and fasting blood glucose. According to the definition of the National Cholesterol Education Program-Adult Treatment Panel III (NCEP-ATP III), individuals having three or more among the five risk factors described above are defined as MetS patients (Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults, 2001). MetS may be caused by the interaction between genetic and environmental factors, such as a multifactorial disease (DeFronzo and Ferrannini, 1991; Groop, 2000; Park et al., 2003). It is known to be the major risk factor for cardiovascular or cerebrovascular diseases such as coronary heart disease (CHD) and stroke, as well as type 2 diabetes (Wannamethee et al., 2005; Janghorbani and Amini, 2012). The prevalence of the MetS patients takes up about 25–35% of adults in the advanced countries (Ervin, 2009). 1 2
Apolipoprotein A5 (ApoA5) plays an important role in the TG metabolism (Dallinga-Thie et al., 2006; Vaessen et al., 2006), despite its low levels. It is highly expressed in the liver and is bound to TG-rich lipoproteins and HDL. It increases the catabolism of TG-rich lipoproteins and inhibits the rate of production of very low-density lipoprotein (Van der Vliet et al., 2001). In animal models, ApoA5 transgenic and knockout mice result in increased and decreased TG levels (Pennacchio et al., 2001; Van der Vliet et al., 2002), suggesting that ApoA5 is inversely associated with TG levels, unlike ApoC3. Thus, the functional defect of ApoA5 is a risk factor for hypertriglyceridemia or vascular diseases. The clones of human ApoA5 gene have been isolated and sequenced. It is located in the chromosome 11q23, *27 kb distal from ApoA4. It is composed of 4 exons and 3 introns and the codes for a 366 amino acid protein. The ApoA5 gene is clustered with ApoA4, ApoC3, and ApoA1 genes (Groenendijk et al., 2001; Pennacchio et al., 2001). So far, several polymorphic sites were found in the vicinity of and within the ApoA5 gene.
Department of Laboratory Medicine, School of Medicine, Jeju National University, Jeju, Republic of Korea. Department of Science Education, Teachers College, Jeju National University, Jeju, Republic of Korea.
130
ASSOCIATION OF THE APOA5 POLYMORPHISMS WITH METS
Table 1. Demographic Characteristics of the Controls and Metabolic Syndrome Patients
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Characteristic
Control (n = 659)
MetS (n = 415)
p-Value
Age, years 48.83 – 10.35 49.47 – 11.58 BMI, kg/m2 23.59 – 2.87 25.91 – 2.99 FBS, mg/dL 86.95 – 10.81 102.84 – 25.68 TG, mg/dL 84.76 – 39.91 211.56 – 163.26 HDL-C, mg/dL 56.84 – 13.41 44.50 – 10.19 SBP, mmHg 117.56 – 11.98 133.61 – 14.06 DBP, mmHg 70.33 – 8.71 81.31 – 10.25 WC, cm 74.31 – 9.61 87.00 – 9.84
0.560 C polymorphic site were classified depending on the presence (T allele, 141 and 21 bp) or absence (C allele, 162 bp) of a restriction enzyme cutting site.
Table 2. Comparison of the Genotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms Between Control Subjects and Patients with Metabolic Syndrome Genotype -1131T>C TT TC CC TT vs. TC+CC TT+TC vs. CC -12,238T>C TT TC CC TT vs. TC+CC TT+TC vs. CC
Controls (n = 659)
MetS (n = 415)
378 (57.4) 233 (35.4) 48 (7.3)
209 (50.4) 152 (36.6) 54 (13.0)
1.000 1.180 2.035 1.326 1.904
(reference) (0.905–1.538) (1.332–3.109) (1.036–1.697) (1.264–2.869)
0.223 0.001 0.028 0.003
1.211 2.084 1.356 1.931
229 (34.7) 300 (45.5) 130 (19.7)
95 (22.9) 212 (51.1) 108 (26.0)
1.000 1.703 2.003 1.794 1.432
(reference) (1.266–2.292) (1.411–2.841) (1.356–2.373) (1.070–1.916)
0.000 0.000 C TT TC CC TT vs. TC+CC TT+TC vs. CC a
Controls (n = 314)
MetS (n = 320)
188 (59.9) 103 (32.8) 23 (7.3)
166 (51.9) 113 (35.3) 41 (12.8)
1.000 1.242 2.019 1.384 1.859
(reference) (0.885–1.744) (1.163–3.505) (1.010–1.896) (1.087–3.179)
0.227 0.014 0.046 0.025
1.297 2.163 1.444 1.930
106 (33.8) 145 (46.2) 63 (20.1)
74 (23.1) 167 (52.2) 79 (24.7)
1.000 1.650 1.796 1.694 1.306
(reference) (1.138–2.391) (1.151–2.802) (1.194–2.403) (0.897–1.901)
0.009 0.010 0.004 0.182
1.619 1.786 1.668 1.308
COR (95% CI)
pa
pa
Statistical power (%)
(0.921–1.828) (1.235–3.787) (1.050–1.985) (1.124–3.314)
0.137 0.007 0.024 0.017
23.0 69.1 51.7 61.8
(1.114–2.353) (1.144–2.788) (1.173–2.372) (0.897–1.908)
0.012 0.011 0.004 0.163
74.5 71.6 83.6 26.3
AOR (95% CI)
Adjusted by age.
Measurements for the risk factors
For the measurements of biochemical concentrations, blood samples were drawn from the subjects, after an overnight fast. The levels of plasma glucose, TG, and HDL-C were determined in duplicates by using enzymatic methods adapted to an automated analyzer (TBA 200FR NEO; Toshiba Medical Systems), according to the procedure recommended by the manufacturer. For anthropometric measurements, body mass index (BMI) was calculated as weight (kg) divided by the square of height (m2). WC was measured using a flexible tape at the smallest horizontal circumference between the costal and margin and the iliac crests. The systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured on the right arm, in the seated position after 10 min of rest. The baseline data of the study subjects are shown in Table 1, which are expressed as mean – standard deviation. Patients with MetS had a significantly higher level of risk factors for MetS development, including BMI, WC, blood pressure,
glucose, and TG, but they had lower level of HDL-C than the control subjects ( p < 0.0001). Statistical analyses
Statistical analyses were conducted using StatsDirect Statistical Software (Version 2.4.4; StatsDirect Ltd.). Clinical profiles between the MetS patients and controls were compared using a Student’s t-test. The chi-square test or Fisher’s exact test was used to compare genotype and haplotype distributions and to assess deviations from the Hardy–Weinberg equilibrium. To evaluate the effect of the ApoA5 genotype on the development of MetS, multivariate logistic regression analysis was performed to derive the odds ratio (OR). The associations between genotypes and the risk factors of MetS were examined using analysis of variance (ANOVA). Haplotypes for two polymorphic loci were estimated using the expectation– maximization algorithm with SNPAlyze (Version 5.1; DYNACOM Co, Ltd.). The level of statistical significance was set at 5%.
Table 4. The Genotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms in the Female Group Genotype -1131T>C TT TC CC TT vs. TC+CC TT+TC vs. CC -12,238T>C TT TC CC TT vs. TC+CC TT+TC vs. CC a
Adjusted by age.
Controls (n = 345)
MetS (n = 95)
COR (95% CI)
190 (55.1) 130 (37.7) 25 (7.2)
43 (45.3) 39 (41.1) 13 (13.7)
1.000 1.326 2.298 1.482 2.029
(reference) (0.814–2.159) (1.088–4.853) (0.939–2.340) (0.995–4.140)
0.262 0.032 0.104 0.062
1.200 1.964 1.332 1.766
123 (35.7) 155 (44.9) 67 (19.4)
21 (22.1) 45 (47.4) 29 (30.5)
1.000 1.700 2.535 1.952 1.823
(reference) (0.962–3.006) (1.342–4.787) (1.146–3.325) (1.093–3.042)
0.072 0.006 0.014 0.025
1.827 2.289 1.990 1.567
pa
AOR (95% CI)
pa
Statistical power (%)
(0.706–2.040) (0.889–4.340) (0.815–2.175) (0.811–3.847
0.501 0.095 0.253 0.152
19.6 53.7 39.1 46.0
(0.997–3.347 (1.149–4.557) (1.126–3.517) (0.901–2.724)
0.051 0.019 0.018 0.112
41.8 80.1 70.8 60.0
ASSOCIATION OF THE APOA5 POLYMORPHISMS WITH METS
133
Table 5. Haplotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms Controls (2n = 1318)
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Haplotype
-1131T>C/–12,238T>C T-T T-C C-T C-C
541 448 217 112
MetS (2n = 830)
(41.0) (34.0) (16.5) (8.5)
249 321 153 107
OR (95% CI)
(30.0) (38.7) (18.4) (12.9)
1.000 1.557 1.532 2.076
(reference) (1.265–1.916) (1.186–1.978) (1.531–2.815)
p-Value
C Polymorphisms According to Gender Male Controls MetS Haplotype (2n = 628) (2n = 640) -1131T>C/–12,238T>C T-T 256 (40.8) 193 T-C 223 (35.5) 252 C-T 101 (16.1) 122 C-C 48 (7.6) 73
(30.2) (39.4) (19.1) (11.4)
Female
OR (95% CI) 1.000 1.499 1.602 2.017
Controls MetS p-Value (2n = 690) (2n = 190)
(reference) (1.156–1.943) (1.160–2.214) (1.340–3.038)
Results
Genotype frequencies of the ApoA5 -1131T>C and -12,238T>C polymorphisms were compared between the control and MetS groups (Table 2). The genotype frequencies for two polymorphisms were in accordance with the Hardy– Weinberg equilibrium. For the -1131T>C polymorphism, the -1131C variant was frequently found in the MetS patients compared with the controls (adjusted OR [AOR] = 2.084, 95% CI = 1.350–3.216, p < 0.001). The dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models were also associated with the increased risk of MetS (AOR = 1.356, 95% CI = 1.054–1.746, p = 0.018; AOR = 1.931, 95% CI = 1.270–2.938, p = 0.002, respectively). When stratified by gender, only the male subgroup showed the same trend as the total subjects, suggesting that the association is male specific (Table 3). The CT and TT genotypes of the ApoA5 -12,238T>C polymorphism were frequently found in the MetS group, compared with the control group (AOR = 1.693, 95% CI = 1.251–2.292, p = 0.001; AOR = 1.938, 95% CI = 1.360– 2.762, p < 0.001). The dominant (CC vs. CT+TT) and recessive (CC+CT vs. TT) models were also associated with the increased MetS susceptibility (AOR = 1.767, 95%
0.002 0.005 0.001
285 225 116 64
(41.3) (32.6) (16.8) (9.3)
55 70 32 33
(28.9) (36.8) (16.8) (17.4)
OR (95% CI) 1.000 1.612 1.429 2.672
(reference) (1.087–2.391) (0.879–2.325) (1.605–4.449)
p-Value
0.021 0.158 0.000
CI = 1.329–0.349, p < 0.001; AOR = 1.387, 95% CI = 1.030– 1.867, p = 0.031, respectively). The male and female subgroups showed almost the same association patterns with the overall subjects, when stratified by gender (Tables 3 and 4). We constructed haplotypes composed of ApoA5 -1131T>C and -12,238T>C polymorphisms (Table 5). The T-C, C-T, and C-C haplotypes were associated with the increased risk for MetS (OR = 1.557, 95% CI = 1.265–1.916, p < 0.001; OR = 1.532, 95% CI = 1.186–1.978, p = 0.001; OR = 2.076, 95% CI = 1.531–1.815, p < 0.001, respectively). When the haplotype data were stratified by gender, the association of each gender was almost the same as that of the total subjects (Table 6). To evaluate the effect of the ApoA5 polymorphisms on the risk factors for MetS pathogenesis, we performed the ANOVA test among the genotypes (Table 7). The -1131T>C and -12,238T>C polymorphisms of the ApoA5 gene were associated with TG and WC levels in the MetS patients ( p < 0.001, p = 0.049, respectively). However, no association was observed in the control group (data not shown). Tables 8 and 9 represent the stratification analysis for the risk factors of MetS in the ApoA5 -1131T>C and -12,238T>C polymorphisms, respectively. In the TC+CC for TT genotype of -1131T>C polymorphism, susceptibility
Table 7. The Effects of the ApoA5 Polymorphisms on Parameters of Metabolic Syndrome Development BMI
WC
-1131T>C TT 27.41 – 2.62 87.31 – 9.50 TC 27.22 – 2.61 86.27 – 10.57 CC 27.43 – 2.82 87.82 – 9.01 p-Value 0.779 0.495 -12,238T>C TT 27.57 – 2.37 84.820 – 9.45 TC 27.07 – 2.59 87.70 – 9.70 CC 27.69 – 2.90 87.53 – 10.26 p-Value 0.087 0.049
SBP
DBP
Glucose
TG
HDL
133.68 – 14.25 81.98 – 9.61 102.63 – 25.78 185.53 – 104.15 45.26 – 11.29 134.14 – 13.46 80.67 – 11.32 105.07 – 27.24 209.28 – 136.79 44.32 – 8.97 131.83 – 15.05 50.50 – 9.456 97.39 – 19.58 318.67 – 312.99 42.02 – 8.50 0.583 0.405 0.166 C Polymorphism on Parameters in the Metabolic Syndrome Group by Stratified Analysis Variables
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
BMI ‡25
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Association of Apolipoprotein A5 Gene Polymorphisms with Metabolic Syndrome in the Korean Population Young Ree Kim1 and Seung-Ho Hong2
Aims: Functional defects of the ApoA5 protein have been identified as risk factors for hypertriglyceridemia, vascular diseases and susceptibility to metabolic syndrome (MetS). These associations are neither strong nor consistent in all populations studied. In this study, we investigated the association between the ApoA5 -1131T>C and -12,238T>C polymorphic loci in Korean patients with MetS. Methods: A total of 1074 subjects, including 415 patients with MetS and 659 healthy control subjects, were enrolled to investigate the affect of ApoA5 polymorphisms on risk of MetS. Genotyping of the ApoA5 polymorphisms was performed by polymerase chain reaction–restriction fragment length polymorphism techniques. Results: The CC genotype and the dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models of the -1131T>C polymorphism were associated with increased MetS susceptibility ( p < 0.001, p = 0.018, and p = 0.002, respectively). The association was malespecific when stratified by gender. With regard to the -12,238T>C polymorphism, the TC and CC genotypes and the dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models were frequently found in the patient group, compared with the control group ( p = 0.001, p < 0.001, p < 0.001, and p = 0.031, respectively). The T-C, C-T, and C-C haplotypes of the ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with an increased risk for MetS ( p < 0.001, p = 0.001, and p < 0.001, respectively). The variant of the ApoA5 -1131T>C polymorphism was also associated with increased triglyceride (TG) levels. Dominant models of ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with the risk components of MetS by the stratification analysis. Conclusion: The -1131C and -12,238C variants and the C-containing haplotypes of ApoA5 -1131T>C and -12,238T>C polymorphisms were associated with higher risk for MetS in the Korean population. The -1131C variant was also associated with the increased level of TG. Introduction
M
etabolic syndrome (MetS) is developed due to abnormality of waist circumference (WC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), blood pressure, and fasting blood glucose. According to the definition of the National Cholesterol Education Program-Adult Treatment Panel III (NCEP-ATP III), individuals having three or more among the five risk factors described above are defined as MetS patients (Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults, 2001). MetS may be caused by the interaction between genetic and environmental factors, such as a multifactorial disease (DeFronzo and Ferrannini, 1991; Groop, 2000; Park et al., 2003). It is known to be the major risk factor for cardiovascular or cerebrovascular diseases such as coronary heart disease (CHD) and stroke, as well as type 2 diabetes (Wannamethee et al., 2005; Janghorbani and Amini, 2012). The prevalence of the MetS patients takes up about 25–35% of adults in the advanced countries (Ervin, 2009). 1 2
Apolipoprotein A5 (ApoA5) plays an important role in the TG metabolism (Dallinga-Thie et al., 2006; Vaessen et al., 2006), despite its low levels. It is highly expressed in the liver and is bound to TG-rich lipoproteins and HDL. It increases the catabolism of TG-rich lipoproteins and inhibits the rate of production of very low-density lipoprotein (Van der Vliet et al., 2001). In animal models, ApoA5 transgenic and knockout mice result in increased and decreased TG levels (Pennacchio et al., 2001; Van der Vliet et al., 2002), suggesting that ApoA5 is inversely associated with TG levels, unlike ApoC3. Thus, the functional defect of ApoA5 is a risk factor for hypertriglyceridemia or vascular diseases. The clones of human ApoA5 gene have been isolated and sequenced. It is located in the chromosome 11q23, *27 kb distal from ApoA4. It is composed of 4 exons and 3 introns and the codes for a 366 amino acid protein. The ApoA5 gene is clustered with ApoA4, ApoC3, and ApoA1 genes (Groenendijk et al., 2001; Pennacchio et al., 2001). So far, several polymorphic sites were found in the vicinity of and within the ApoA5 gene.
Department of Laboratory Medicine, School of Medicine, Jeju National University, Jeju, Republic of Korea. Department of Science Education, Teachers College, Jeju National University, Jeju, Republic of Korea.
130
ASSOCIATION OF THE APOA5 POLYMORPHISMS WITH METS
Table 1. Demographic Characteristics of the Controls and Metabolic Syndrome Patients
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Characteristic
Control (n = 659)
MetS (n = 415)
p-Value
Age, years 48.83 – 10.35 49.47 – 11.58 BMI, kg/m2 23.59 – 2.87 25.91 – 2.99 FBS, mg/dL 86.95 – 10.81 102.84 – 25.68 TG, mg/dL 84.76 – 39.91 211.56 – 163.26 HDL-C, mg/dL 56.84 – 13.41 44.50 – 10.19 SBP, mmHg 117.56 – 11.98 133.61 – 14.06 DBP, mmHg 70.33 – 8.71 81.31 – 10.25 WC, cm 74.31 – 9.61 87.00 – 9.84
0.560 C polymorphic site were classified depending on the presence (T allele, 141 and 21 bp) or absence (C allele, 162 bp) of a restriction enzyme cutting site.
Table 2. Comparison of the Genotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms Between Control Subjects and Patients with Metabolic Syndrome Genotype -1131T>C TT TC CC TT vs. TC+CC TT+TC vs. CC -12,238T>C TT TC CC TT vs. TC+CC TT+TC vs. CC
Controls (n = 659)
MetS (n = 415)
378 (57.4) 233 (35.4) 48 (7.3)
209 (50.4) 152 (36.6) 54 (13.0)
1.000 1.180 2.035 1.326 1.904
(reference) (0.905–1.538) (1.332–3.109) (1.036–1.697) (1.264–2.869)
0.223 0.001 0.028 0.003
1.211 2.084 1.356 1.931
229 (34.7) 300 (45.5) 130 (19.7)
95 (22.9) 212 (51.1) 108 (26.0)
1.000 1.703 2.003 1.794 1.432
(reference) (1.266–2.292) (1.411–2.841) (1.356–2.373) (1.070–1.916)
0.000 0.000 C TT TC CC TT vs. TC+CC TT+TC vs. CC a
Controls (n = 314)
MetS (n = 320)
188 (59.9) 103 (32.8) 23 (7.3)
166 (51.9) 113 (35.3) 41 (12.8)
1.000 1.242 2.019 1.384 1.859
(reference) (0.885–1.744) (1.163–3.505) (1.010–1.896) (1.087–3.179)
0.227 0.014 0.046 0.025
1.297 2.163 1.444 1.930
106 (33.8) 145 (46.2) 63 (20.1)
74 (23.1) 167 (52.2) 79 (24.7)
1.000 1.650 1.796 1.694 1.306
(reference) (1.138–2.391) (1.151–2.802) (1.194–2.403) (0.897–1.901)
0.009 0.010 0.004 0.182
1.619 1.786 1.668 1.308
COR (95% CI)
pa
pa
Statistical power (%)
(0.921–1.828) (1.235–3.787) (1.050–1.985) (1.124–3.314)
0.137 0.007 0.024 0.017
23.0 69.1 51.7 61.8
(1.114–2.353) (1.144–2.788) (1.173–2.372) (0.897–1.908)
0.012 0.011 0.004 0.163
74.5 71.6 83.6 26.3
AOR (95% CI)
Adjusted by age.
Measurements for the risk factors
For the measurements of biochemical concentrations, blood samples were drawn from the subjects, after an overnight fast. The levels of plasma glucose, TG, and HDL-C were determined in duplicates by using enzymatic methods adapted to an automated analyzer (TBA 200FR NEO; Toshiba Medical Systems), according to the procedure recommended by the manufacturer. For anthropometric measurements, body mass index (BMI) was calculated as weight (kg) divided by the square of height (m2). WC was measured using a flexible tape at the smallest horizontal circumference between the costal and margin and the iliac crests. The systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured on the right arm, in the seated position after 10 min of rest. The baseline data of the study subjects are shown in Table 1, which are expressed as mean – standard deviation. Patients with MetS had a significantly higher level of risk factors for MetS development, including BMI, WC, blood pressure,
glucose, and TG, but they had lower level of HDL-C than the control subjects ( p < 0.0001). Statistical analyses
Statistical analyses were conducted using StatsDirect Statistical Software (Version 2.4.4; StatsDirect Ltd.). Clinical profiles between the MetS patients and controls were compared using a Student’s t-test. The chi-square test or Fisher’s exact test was used to compare genotype and haplotype distributions and to assess deviations from the Hardy–Weinberg equilibrium. To evaluate the effect of the ApoA5 genotype on the development of MetS, multivariate logistic regression analysis was performed to derive the odds ratio (OR). The associations between genotypes and the risk factors of MetS were examined using analysis of variance (ANOVA). Haplotypes for two polymorphic loci were estimated using the expectation– maximization algorithm with SNPAlyze (Version 5.1; DYNACOM Co, Ltd.). The level of statistical significance was set at 5%.
Table 4. The Genotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms in the Female Group Genotype -1131T>C TT TC CC TT vs. TC+CC TT+TC vs. CC -12,238T>C TT TC CC TT vs. TC+CC TT+TC vs. CC a
Adjusted by age.
Controls (n = 345)
MetS (n = 95)
COR (95% CI)
190 (55.1) 130 (37.7) 25 (7.2)
43 (45.3) 39 (41.1) 13 (13.7)
1.000 1.326 2.298 1.482 2.029
(reference) (0.814–2.159) (1.088–4.853) (0.939–2.340) (0.995–4.140)
0.262 0.032 0.104 0.062
1.200 1.964 1.332 1.766
123 (35.7) 155 (44.9) 67 (19.4)
21 (22.1) 45 (47.4) 29 (30.5)
1.000 1.700 2.535 1.952 1.823
(reference) (0.962–3.006) (1.342–4.787) (1.146–3.325) (1.093–3.042)
0.072 0.006 0.014 0.025
1.827 2.289 1.990 1.567
pa
AOR (95% CI)
pa
Statistical power (%)
(0.706–2.040) (0.889–4.340) (0.815–2.175) (0.811–3.847
0.501 0.095 0.253 0.152
19.6 53.7 39.1 46.0
(0.997–3.347 (1.149–4.557) (1.126–3.517) (0.901–2.724)
0.051 0.019 0.018 0.112
41.8 80.1 70.8 60.0
ASSOCIATION OF THE APOA5 POLYMORPHISMS WITH METS
133
Table 5. Haplotype Frequencies of the ApoA5 -1131T>C and -12,238T>C Polymorphisms Controls (2n = 1318)
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
Haplotype
-1131T>C/–12,238T>C T-T T-C C-T C-C
541 448 217 112
MetS (2n = 830)
(41.0) (34.0) (16.5) (8.5)
249 321 153 107
OR (95% CI)
(30.0) (38.7) (18.4) (12.9)
1.000 1.557 1.532 2.076
(reference) (1.265–1.916) (1.186–1.978) (1.531–2.815)
p-Value
C Polymorphisms According to Gender Male Controls MetS Haplotype (2n = 628) (2n = 640) -1131T>C/–12,238T>C T-T 256 (40.8) 193 T-C 223 (35.5) 252 C-T 101 (16.1) 122 C-C 48 (7.6) 73
(30.2) (39.4) (19.1) (11.4)
Female
OR (95% CI) 1.000 1.499 1.602 2.017
Controls MetS p-Value (2n = 690) (2n = 190)
(reference) (1.156–1.943) (1.160–2.214) (1.340–3.038)
Results
Genotype frequencies of the ApoA5 -1131T>C and -12,238T>C polymorphisms were compared between the control and MetS groups (Table 2). The genotype frequencies for two polymorphisms were in accordance with the Hardy– Weinberg equilibrium. For the -1131T>C polymorphism, the -1131C variant was frequently found in the MetS patients compared with the controls (adjusted OR [AOR] = 2.084, 95% CI = 1.350–3.216, p < 0.001). The dominant (TT vs. TC+CC) and recessive (TT+TC vs. CC) models were also associated with the increased risk of MetS (AOR = 1.356, 95% CI = 1.054–1.746, p = 0.018; AOR = 1.931, 95% CI = 1.270–2.938, p = 0.002, respectively). When stratified by gender, only the male subgroup showed the same trend as the total subjects, suggesting that the association is male specific (Table 3). The CT and TT genotypes of the ApoA5 -12,238T>C polymorphism were frequently found in the MetS group, compared with the control group (AOR = 1.693, 95% CI = 1.251–2.292, p = 0.001; AOR = 1.938, 95% CI = 1.360– 2.762, p < 0.001). The dominant (CC vs. CT+TT) and recessive (CC+CT vs. TT) models were also associated with the increased MetS susceptibility (AOR = 1.767, 95%
0.002 0.005 0.001
285 225 116 64
(41.3) (32.6) (16.8) (9.3)
55 70 32 33
(28.9) (36.8) (16.8) (17.4)
OR (95% CI) 1.000 1.612 1.429 2.672
(reference) (1.087–2.391) (0.879–2.325) (1.605–4.449)
p-Value
0.021 0.158 0.000
CI = 1.329–0.349, p < 0.001; AOR = 1.387, 95% CI = 1.030– 1.867, p = 0.031, respectively). The male and female subgroups showed almost the same association patterns with the overall subjects, when stratified by gender (Tables 3 and 4). We constructed haplotypes composed of ApoA5 -1131T>C and -12,238T>C polymorphisms (Table 5). The T-C, C-T, and C-C haplotypes were associated with the increased risk for MetS (OR = 1.557, 95% CI = 1.265–1.916, p < 0.001; OR = 1.532, 95% CI = 1.186–1.978, p = 0.001; OR = 2.076, 95% CI = 1.531–1.815, p < 0.001, respectively). When the haplotype data were stratified by gender, the association of each gender was almost the same as that of the total subjects (Table 6). To evaluate the effect of the ApoA5 polymorphisms on the risk factors for MetS pathogenesis, we performed the ANOVA test among the genotypes (Table 7). The -1131T>C and -12,238T>C polymorphisms of the ApoA5 gene were associated with TG and WC levels in the MetS patients ( p < 0.001, p = 0.049, respectively). However, no association was observed in the control group (data not shown). Tables 8 and 9 represent the stratification analysis for the risk factors of MetS in the ApoA5 -1131T>C and -12,238T>C polymorphisms, respectively. In the TC+CC for TT genotype of -1131T>C polymorphism, susceptibility
Table 7. The Effects of the ApoA5 Polymorphisms on Parameters of Metabolic Syndrome Development BMI
WC
-1131T>C TT 27.41 – 2.62 87.31 – 9.50 TC 27.22 – 2.61 86.27 – 10.57 CC 27.43 – 2.82 87.82 – 9.01 p-Value 0.779 0.495 -12,238T>C TT 27.57 – 2.37 84.820 – 9.45 TC 27.07 – 2.59 87.70 – 9.70 CC 27.69 – 2.90 87.53 – 10.26 p-Value 0.087 0.049
SBP
DBP
Glucose
TG
HDL
133.68 – 14.25 81.98 – 9.61 102.63 – 25.78 185.53 – 104.15 45.26 – 11.29 134.14 – 13.46 80.67 – 11.32 105.07 – 27.24 209.28 – 136.79 44.32 – 8.97 131.83 – 15.05 50.50 – 9.456 97.39 – 19.58 318.67 – 312.99 42.02 – 8.50 0.583 0.405 0.166 C Polymorphism on Parameters in the Metabolic Syndrome Group by Stratified Analysis Variables
Downloaded by UNIVERSITY OF VIRGINIA E-PACKAGE from online.liebertpub.com at 10/06/17. For personal use only.
BMI ‡25
