Journal of Human Hypertension (2007) 21, 673–682 & 2007 Nature Publishing Group All rights reserved 0950-9240/07 $30.00 www.nature.com/jhh
ORIGINAL ARTICLE
Polymorphisms of angiotensinogen and angiotensin-converting enzyme associated with lower extremity arterial disease in the Health, Aging and Body Composition study R Li1, B Nicklas2, M Pahor3, A Newman4, K Sutton-Tyrrell5, T Harris6, E Lakatta7, DC Bauer8, J Ding2, S Satterfield9 and SB Kritchevsky2 1
Department of Preventive Medicine, Center for Genomics and Bioinformatics, University of Tennessee Health Science Center, Memphis, TN, USA; 2Department of Internal Medicine, Wake Forest University School of Medicine, Winston-Salem, NC, USA; 3Department of Aging and Geriatric Research, University of Florida College of Medicine, Gainesville, FL, USA; 4Division of Geriatric Medicine, University of Pittsburgh, PA, USA; 5 Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA, USA; 6 Geriatric Epidemiology, National Institute on Aging, National Institutes of Health, USA; 7Laboratory of Cardiovascular Science and Cardiovascular Function Section, National Institute on Aging, National Institutes of Health, USA; 8Department of Epidemiology and Biostatistics, University of California, San Francisco, CA, USA and 9Department of Preventive Medicine, University of Tennessee Health Science Center, Memphis, TN, USA
The role of renin–angiotensin system (RAS) genes on the risk of lower extremity arterial disease (LEAD) in elderly people remains unclear. We assessed the relationship of genetic polymorphisms in RAS: G-6A, T174M and M235T of the angiotensinogen (AGT) gene, and the angiotensin-converting enzyme insertion/deletion (ACE_I/D) variant to the risk of LEAD in the Health, Aging and Body Composition (Health ABC) Study. This analysis included 1228 black and 1306 white men and women whose age ranged between 70 and 79 years at the study enrollment. LEAD was defined as ankle-arm index (AAI) o0.9. Genotype–phenotype associations were estimated by regression analyses with and without adjustment for established cardiovascular disease (CVD) risk factors. The proportion of LEAD was significantly higher in black (21.1%) than that in white elderly people (10.1%, Po0.0001). The distribution of
AGT polymorphisms was also significantly different between black and white participants. There was no statistically significant association between the selected RAS genetic variants and LEAD after adjustment for age, antihypertensive medications, lipid-lowering medication, pack-year smoking, body mass index, low-density lipoprotein cholesterol, and prevalent diabetes and coronary heart disease. However, A-T haplotype of G6A and M235T interacting with homozygous ACE_II (b ¼ 1.07, P ¼ 0.006) and with ACE inhibitors (b ¼ 1.03, P ¼ 0.01) significantly decreased the risk of LEAD in white but not in black participants after adjustment for the selected CVD risk factors. In conclusion, the study observed a gene–gene and gene–drug interaction for LEAD in the white elderly. Journal of Human Hypertension (2007) 21, 673–682; doi:10.1038/sj.jhh.1002198; published online 12 April 2007
Keywords: angiotensinogen polymorphisms; angiotensin-converting enzyme insertion/deletion variant; lower extremity
disease (LEAD); elderly; race
Introduction The renin–angiotensin system (RAS) regulates blood pressure (BP) by its effects on vascular tone, renal Correspondence: Dr R Li, Department of Preventive Medicine, University of Tennessee Health Science Center, 66 N Pauline, Suite 633, Memphis, TN 38163, USA. E-mail:
[email protected] Received 4 November 2006; revised 8 February 2007; accepted 5 March 2007; published online 12 April 2007
haemodynamics, and sodium and volume homeostasis. The principal vasoactive component of the RAS is angiotensin II, which exerts powerful vasoconstrictive effects.1–3 Angiotensin II is generated in two steps: (1) renin cleaves angiotensinogen (AGT) to angiotensin I; and (2) angiotensin-converting enzyme (ACE) converts angiotensin I to angiotensin II. The AGT polymorphism M235T corresponds to a change from methionine to threonine at position 235.4,5 The polymorphism T174M in
AGT, ACE and LEAD R Li et al 674
a tight linkage disequilibrium with the M235T variant4 encodes methionine instead of threonine. Studies4,6 have found higher plasma AGT levels among 235T allele carriers, but no association between T174M variant and AGT levels. The G-6A variant in the promoter region of AGT gene is in linkage disequilibrium with M235T and T174M, which causes a change in AGT gene expression, thereby altering plasma AGT levels.7,8 The ACE insertion/deletion polymorphism (ACE_I/D), the presence or absence of a 287 base pair DNA fragment in intron 16 of the gene, affects plasma ACE level, with a higher plasma ACE level in D allele carriers.9–11 Thus, people who carry susceptibility polymorphisms of RAS may experience chronically unbalanced vasoconstriction and vasorelaxation. The unbalanced vascular tones increase the arterial stiffness and, consequently, increase the risk of development of lower extremity arterial disease (LEAD) or other vascular diseases. Previous reports on the association of essential hypertension12–14 and cardiovascular diseases15,16 with G-6A, T174M, M235T and ACE are inconsistent, possibly due to small sample size or unrecognized confounders or effect modifiers. In addition, the relationship between RAS genes and LEAD in Europeans and Asians17–19 is also inconsistent. To our knowledge, no study has yet examined the association between variations of RAS genes and LEAD in elderly and black participants. In this study, we estimated the relationship of G-6A, T174M, M235T of the AGT gene and ACE_I/D to LEAD and tested gene–gene and gene–covariate interaction in black and white men and women who were 70–79 years old and enrolled in the Health, Aging and Body Composition (Health ABC) Study.
ducted using the baseline data of the remaining 1228 black and 1306 white participants. All participants signed a written informed consent approved by the institutional review boards of the University of Pittsburgh and University of Tennessee.
Materials and methods
Candidate polymorphism assays
Participants
The Health ABC study is a population-based, prospective study of the impact of changes in weight and body composition on age-related physiological and functional changes. Participants aged 70–79 years were recruited from March 1997 to July 1998 at two field centres located in Pittsburgh, PA, and Memphis, TN. They were drawn from a random sample of Medicare beneficiaries residing in ZIP codes from the metropolitan areas surrounding Pittsburgh and Memphis. Eligible subjects reported no difficulty walking one-quarter of a mile, climbing 10 steps, or performing basic activities of daily living. They also had to be free of life-threatening illness and to plan to remain in the area for at least 3 years. The study consisted of 3075 men (48.4%) and women, of whom 41.7% were black. Participants with missing genotype data (n ¼ 102) or with allelic distribution under Hardy–Weinberg disequilibrium (439 white women from Memphis) were excluded from the analyses. Thus, this analysis was conJournal of Human Hypertension
Definition of LEAD
All participants had their ankle-arm index (AAI) measured at the baseline clinical visit (1997–1998). AAI is a noninvasive procedure for detecting LEAD. Arterial systolic blood pressure (SBP) was measured by a hand-held, 8-MHz Doppler probe placed directly over the artery and a conventional mercury sphygmomanometer. The participants were in supine position on an examination table, and after 5 min of rest, standard adult-size BP cuffs were applied to the right arm and to each ankle (with the lower end of the bladder within 3 cm of the malleoli). After palpation of the brachial and posterior tibial arteries, ultrasound gel was applied, and a Doppler stethoscope (8 MHz, Huntleigh Technology, Inc., Manalapan, NJ, USA) and a standard mercury manometer were used to assess SBP in the right brachial artery and in each posterior tibial artery in rapid succession. Measurements have been shown to be reliable between observers, stable over time, and highly correlated between left and right legs.20 Means of the first and second SBP measurements for each leg and right arm were used to attain AAI. AAI was defined as the lowest ratio of SBP of either the right ankle to the right upper-arm or the left ankle to the right upper-arm. According to previous reports of abnormal AAI associated with preclinical cardiovascular disease,21 AAI p0.9 in either leg was used to define LEAD.
Genomic DNA was extracted and purified from buffy coat prepared from whole blood using the PUREGENE DNA Purification Kit (Gentra System, Inc., Minneapolis, MN, USA), and stored at 801C until analysis. The ACE insertion/deletion polymorphism in intron 16 of the ACE gene was determined using polymerase chain reaction (PCR) amplification with subsequent visualization of PCR products on 2% agarose gels by electrophoresis. The sequences of the sense and antisense primers were 50 -CTGGA GACCACTCCCATCCTTTCT-30 and 50 GATGTGGCC ATCACATTCGTCAGAT-30 , respectively. The insertion allele (I) was detected as a 490-base pair band, and the deletion allele (D) was visualized as a 190base pair band. The PCR products were visualized independently by two laboratory technicians, and genotypes that were not scored identically by both technicians were re-analyzed. Because the D allele in heterozygous samples is preferentially amplified,22 all samples were re-amplified using a primer pair that recognizes an insertion-specific sequence
AGT, ACE and LEAD R Li et al
(50 -TGGGACCACAGCGCCCGCCACTAC-30 ; 50 -TCGC CAGCCCTCCCATGCCCATAA-30 ). The reaction yields a 335-base pair band only in the presence of an I allele and no product in DD homozygotes. This procedure identified 45 samples (1.5%) with the ID genotype that had been misclassified as DD with the initial primers. G-6A, T174M and M235T polymorphisms of AGT were genotyped using the MassARRAY system (SEQUENOM, Inc., San Diego, CA, USA). PCR reactions were performed in a total volume of 5 ml with 10 ng of genomic DNA, 1 PCR buffer (Qiagen, Valencia, CA, USA), 2.5 mM MgCl2 (Qiagen), 0.1 U of Hot StarTaq polymerase (Qiagen), 200 mM dNTP (Invitrogen, Carlsbad, CA, USA), and 200 nM of each primer. The extension reactions were performed in a total volume of 9 ml with 50 mM dNTP/dideoxynucleotide phosphate (ddNTP) each, 0.063 U/ml Thermo Sequenase (both from SEQUENOM, Inc.), and 600 nM extension primers. Genotyping was carried out using proprietary software (SpectroTYPER, SEQUENOM, Inc., San Diego, CA, USA). Because the genotyping methods were unable to distinguish gametic phase of polymorphisms, eight (23) possible haplotypes of diallelic polymorphisms G-6A, T174M, and M235T and four (22) possible haplotypes of G-6A and M235T were estimated using expectation-maximization (EM) algorithm. Covariates
Covariates, which were collected at baseline, included sociodemographic variables (age, sex, race, and study site), BP, comorbidity (prevalent coronary heart disease (PCHD), cerebrovascular disease, congestive heart failure, and diabetes), physical, behavioural, and biological parameters (pack-year smoking, Body mass index (BMI), low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol), and antihypertensive and/or lipid-lowering medications. Race was determined by participant self-identification as either black or white in answering the question ‘How would you describe your primary racial or ethnic group’? SBP or diastolic blood pressure (DBP) was an average of two measurements of sitting BP, using a conventional mercury sphygmomanometer on either arm after a 5-min seated rest. Participants with SBPX140 mm Hg, DBPX90 mm Hg, or those using antihypertensive medications in the previous 2 weeks were considered to have elevated BP. PCHD was defined based on self-report of coronary artery bypass grafting surgery (CABG), percutaneous transluminal coronary angioplasty (PTCA), carotid endarterectomy, doctor-diagnosed myocardial infarction (MI) or angina, and use of anti-anginal medications. Prevalent cerebrovascular disease (PCEVD) was identified by self-reported history of transient ischemic attack (TIA) or stroke. Diabetes was determined by self-report of doctor-diagnosed diabetes, fasting glucoseX126 mg/dl, or use of
675
antidiabetic medications. Prevalent congestive heart failure (PCHF) was also a self-reported history of doctor-diagnosed CHF. The information of pack-year smoking was collected from questionnaire. BMI was calculated as weight (kilograms) divided by height (meters) squared using objective measures. Total cholesterol and HDL cholesterol were measured by a colorimetric technique on a Johnson & Johnson Vitros 950 analyzer (Raritan, NJ, USA). LDL levels were calculated by use of the Friedewald equation. Medications taken in the previous 2 weeks were brought in, recorded, and coded according to the Iowa Drug Information System.23 Data analysis
The SAS/Genetics v9.124. was used to test Hardy– Weinberg equilibrium and linkage disequilibrium of selected candidate polymorphisms, and to estimate the haplotype frequency of G-6A, T174M, and M235T polymorphisms of AGT with unknown gametic phase. Maximum-likelihood estimates of haplotype frequency were computed using an EM algorithm. Only common haplotypes with frequencies greater than 5% were presented in the results section. The SAS/STAT was used to compute means of continuous measures, and proportions of categorical measures of variables of interest. Covariate–phenotype, LEAD, and covariate–genotype/haplotype associations were tested separately by race and sex group. Since there was no sex difference in gene–LEAD association, the final analyses were stratified by race only. Covariates that were significantly associated with both genetic variants and LEAD were included in the final logistic regression models. Gene–gene and gene– covariate interactions were also estimated. Because three polymorphisms of the AGT gene and the ACE_I/D variant were included in this genetic association study, we applied the Holm method25,26 to control the family-wise error rate, the probability of making one or more type I errors among all the hypotheses when performing multiple pair-wise tests, without assuming independence. P-values from single polymorphism–LEAD association tests (with or without adjustment for covariates) were considered as raw P-values for calculating adjusted P-values for multiple testing. The adjusted P-values were calculated using the SAS MULTTEST procedure.24 In addition to the adjustment for multiple testing, we also conducted haplotype association analyses using trend regression to avoid multiple testing and to increase statistical power.27 The haplotype association analyses were restricted in haplotypes of two polymorphisms (G-6A and M235T), because of a high linkage disequilibrium between G-6A and T174M (r ¼ 0.97), and T174M associated with cardiovascular phenotype most likely through the partial disequilibrium with alleles at codon 235.4 This two-polymorphism haplotype association analysis further increased the statistical power. Journal of Human Hypertension
AGT, ACE and LEAD R Li et al 676
Table 1 Mean or proportion of participants’ characteristics by race and LEAD Black participants
Pa
White participants
LEAD+ (n ¼ 234) LEAD (n ¼ 872) Total
Pb
LEAD+ (n ¼ 124) LEAD (n ¼ 1133) Total
Pb
ACE II (%) ID (%)
21.4 47.4
16.7 47.8
17.7 47.7
0.20
23.4 46.8
18.4 49.4
18.9 49.2
0.40
0.35
G-6A AA (%) GA (%)
66.5 32.2
68.5 27.6
68.1 28.5
0.07
18.7 51.2
19.4 49.5
19.3 49.6
0.93
o0.0001
M235T TT (%) MT (%)
63.1 35.2
65.9 29.3
65.3 30.6
0.04
18.7 48.0
18.7 49.4
18.7 49.3
0.94
o0.0001
T174M MM (%) TM (%)
0.4 13.7
0.7 12.7
0.7 12.9
0.84
3.3 17.1
1.5 20.9
1.7 20.6
0.25
o0.0001
H1c A-T (%) G-M (%)
80.6 17.5
80.5 17.7
80.5 17.6
0.96 0.93
42.7 56.0
43.0 55.3
43.0 55.4
0.93 0.83
o0.0001 o0.0001
H2d A-M-T (%) A-T-T (%) G-T-M (%)
7.1 73.5 17.3
6.9 73.5 17.6
7.0 73.5 17.6
0.89 0.96 0.87
11.7 30.0 56.0
11.9 31.1 55.2
11.9 31.1 55.3
0.93 0.98 0.81
o0.0001 o0.0001 o0.0001
42.7 68.2 84.2 33.8 31.7 12.6 3.9 68.0 15.4 35.5 20.9 34.6 18.0 74.1 144.3 73.7 28.0 127.4 56.0 24.0
42.8 51.4 73.9 22.4 17.6 6.9 2.2 60.3 10.1 27.1 16.0 33.5 9.5 73.2 136.8 73.5 28.8 123.7 57.5 14.1
42.8 54.9 76.0 24.8 20.6 8.1 2.9 62.7 11.0 29.4 17.3 34.5 11.3 73.4 139.4 73.6 28.7 124.1 57.3 16.8
0.99 o0.001 0.001 o0.001 o0.001 0.005 0.16 0.03 0.03 0.01 0.07 0.7 o0.001 o0.001 o0.001 0.87 0.02 0.16 0.22 o0.001
70.2 78.0 83.1 21.0 33.9 13.0 5.7 67.7 21.8 25.0 22.6 21.0 18.6 74.5 141.8 69.5 26.1 121.2 48.8 33.7
69.5 62.2 61.3 14.1 21.3 5.8 1.4 45.2 13.9 16.8 12.7 16.2 14.6 73.6 132.5 70.0 26.6 119.7 52.3 19.7
69.5 63.8 63.5 14.7 22.5 6.5 1.8 47.5 14.6 17.6 14.0 16.8 15.0 73.8 134.3 69.7 26.4 120.4 49.8 24.6
0.87 o0.001 o0.001 0.04 0.001 0.002 0.005 o0.0001 0.02 0.02 0.002 0.2 0.24 o0.001 o0.001 0.64 0.15 0.59 0.01 o0.001
o0.0001 o0.0001 o0.0001 o0.0001 0.53 0.003 0.06 o0.0001 0.008 o0.0001 0.02 o0.0001 0.01 0.003 o0.0001 o0.0001 o0.0001 0.01 o0.0001 o0.0001
Male (%) Ever SMK (%) HTN (%) Diabetes (%) PCHD (%) PCEVD (%) PCHF (%) HTN Med (%) b-Blocker (%) CCB (%) ACE_INH (%) Diuretics (%) Statin (%) Age (years) SBP (mm Hg) DBP (mm Hg) BMI (kg/m2) LDL-C (mg/dl) HDL-C (mg/dl) PKYR Smoking
Abbreviations: ACE, angiotensin-converting enzyme; ACE_INH, ACE inhibitor; BMI, body mass index; CCB, calcium channel blocker; HDL-C, high-density lipoprotein cholesterol; HTN, hypertension; HTN Med, antihypertensive medications; LDL-C, low-density lipoprotein cholesterol; LEAD, lower extremity arterial disease; PCEVD, prevalent cerebrovascular disease; PCHD, prevalent coronary heart disease; PCHF, prevalent congestive heart failure; PKYR, pack-year smoking; SBP and DBP, systolic and diastolic blood pressure; SMK, smoking; Statin, lipid-lowering medication. a P-value for the comparison between black and white participants. b P-value for the comparison between LEAD+ and LEAD. c H1: haplotypes of G-6A and M235T. d H2: haplotypes of G-6A, T174M and M235T
Results Participants’ characteristics
Participants’ genotype and haplotype frequencies by race and LEAD are presented in Table 1. ACE genotype frequencies were not significantly different between black and white participants. However, other genotype and haplotype distributions differ Journal of Human Hypertension
significantly by race (Po0.001 in the last column of Table 1). The higher frequency of homozygote AA of G-6A (0.681), TT of M235T (0.653), and lower frequency of homozygote MM of T174M (0.007) were observed in black compared with their white counterparts (AA 0.193, TT 0.187, and MM 0.017). Black participants were more likely to have A-T haplotype (0.805), whereas whites were more likely
AGT, ACE and LEAD R Li et al 677 30 P = 0.07
P = 0.04
P = 0.03
25 20 %
15 10 5 0 DD ID II ACE
GGGA AA G-6A
TT TM MM T174M
MM MT TT M235T
0
1 2 A-T
Blacks 30
25
20
%
to have G-M haplotype (0.554) of G-6A and M235T polymorphisms of the AGT gene. The genotype and haplotype distributions were not different by LEAD in both black and white participants, except for the distribution of M235T in black participants. The LEAD cases in whites were more likely to carry T allele or heterozygote of M235T (P ¼ 0.04). As expected, there was no difference in the distribution of autosomal variants by sex (data not shown). The distributions of other cardiovascular disease risk factors by race and LEAD are also presented in Table 1. Black participants were slightly younger (mean age of 73.4 years) than white participants (mean age of 73.8 years, P ¼ 0.003). However, the black participants had a significantly poorer health status indicated by their higher SBP and DBP, higher BMI, LDL cholesterol level, higher prevalence of hypertension, and proportion of using antihypertensive medications (but proportion of b-blocker users was lower in black than that in white participants), higher prevalence of diabetes and/or PCEVD compared with the health status of their white counterparts. The prevalence of LEAD was also statistically significantly higher in black (21.1%) than that in white participants (10.1%, Po0.001). On the other hand, white participants were more likely to have smoked and had lower HDL cholesterol level. The proportion of using statin was higher in white than in black participants. Compared to non-LEAD cases, LEAD cases had significantly higher BP, PCHD, PCEVD, and use of antihypertensive medications in both black and white participants. The LEAD cases were more likely to be smokers and statin users.
15
10
5
0 DD ID II ACE
GG GA AA G-6A
TT TM MM T174M
MM MT TT M235T
0
1 2 A-T
Whites
Figure 1 Proportion of lower extremity arterial disease by genotype and A-T haplotype (0 ¼ no A-T haplotype, 1 ¼ heterozygote with one A-T and 2 ¼ homozygote with a pair of A-T) of G-6A and M235T. Upper panel is for black and bottom panel for white participants.
Haplotype–Phenotype association Genotype and phenotype associations
Genotype–phenotype associations are shown in Figure 1 and Table 2. The proportion of LEAD was higher in black participants who carry the -6A compared to GG homozygote (P ¼ 0.07) and 235T compared to MM homozygote (P ¼ 0.04) (Figure 1) before adjustment for other CVD risk factors. Log odds ratios (ORs) were 1.15 (P ¼ 0.03) for the TT genotype and 1.25 (P ¼ 0.01) for the MT genotype of M235T associated with LEAD in black participants before adjustment for other CVD risk factors (Table 2). After adjustment for the selected CVD risk factors, the association was attenuated (Log OR ¼ 0.89, P ¼ 0.07 for TT and Log OR ¼ 1.00, P ¼ 0.048 for MT). Moreover, this association was no longer statistically significant (P ¼ 0.44 for TT and P ¼ 0.33 for MT) after adjustment for multiple testing (Table 2). In black subjects, the ACE_II genotype was associated with the risk of LEAD after adjustment for the selected CVD risk factors (Log OR ¼ 0.49, P ¼ 0.03), but was not statistically significant (P ¼ 0.25) after adjustment for multiple testing (Table 2). In white participants, none of the selected RAS polymorphisms was significantly associated with LEAD (Figure1 and Table 2).
The haplotype association was estimated according to different modes of inheritance: dominant (with at least one copy of A-T haplotype of G-6A and M235T polymorphisms), recessive (with both A-T haplotype on the pair of chromosomes) and additive (the probability of having A-T haplotype as 0, 0.5, and 1). Before adjustment for the selected CVD risk factors, the A-T haplotype significantly increased the risk of LEAD in black participants (P ¼ 0.03 in Figure 1 and Log OR ¼ 1.04, P ¼ 0.05 in Table 3) with a dominant mode of inheritance. After adjustment for the CVD risk factors, the association was no longer statistically significant (Log OR ¼ 0.72, P ¼ 0.18 in Table 3). No A-T haplotype and LEAD association was observed in white participants in this study. Gene–gene or Gene–covariate interaction
The study observed an interaction between ACE_II and A-T haplotype of G-6A and M235T polymorphisms for the risk of LEAD in whites but not in black participants (Table 4). The interaction was significant (b ¼ 1.826, P ¼ 0.009 before and b ¼ 2.137, P ¼ 0.006 after adjustment for the selected cardiovascular disease risk factors) towards the reduction of the LEAD risk in white elderly people. No ACE_II Journal of Human Hypertension
AGT, ACE and LEAD R Li et al 678
Table 2 Genetic variants associated with lower extremity arterial disease Adjusted for covariatesa
Unadjusted for covariates P1
P2
P1
P2
0.37 0.12
0.08 0.48
0.46 1.00
0.49 0.07
0.03 0.70
0.25 1.00
0.51 0.67
0.22 0.12
0.87 0.60
0.26 0.35
0.54 0.43
1.00 1.00
T174M MM TM
13.38 0.07
0.98 0.79
1.00 1.00
13.43 0.09
0.98 0.71
1.00 1.00
M235T TT MT
1.15 1.25
0.03 0.01
0.19 0.08
0.89 1.00
0.07 0.048
0.44 0.33
0.32 0.02
0.22 0.92
1.00 1.00
0.37. 0.09
0.19 0.71
1.00 1.00
0.04 0.11
0.88 0.60
1.00 1.00
0.09 0.04
0.76 0.88
1.00 1.00
T174M MM TM
0.73 0.22
0.19 0.39
1.00 1.00
1.01 0.24
0.09 0.37
0.71 1.00
M235T TT MT
0.01 0.03
0.98 0.87
1.00 1.00
0.10 0.10
0.74 0.66
1.00 1.00
b Black participants ACE II ID G-6A AA GA
White participants ACE II ID G-6A AA GA
b
Abbreviations: ACE, angiotensin-converting enzyme. a Adjusted for age, pack-year smoking, body mass index, diabetes, low-density lipoprotein cholesterol, coronary heart disease, and use of antihypertensive medication and lipid-lowering medication (statin). P1: P-value for b before adjustment for multiple testing; P2: P-value for b after adjustment for multiple testing.
Table 3 Haplotype of G-6A and M235T associated with lower extremity arterial disease Unadjusted
Adjusteda
b
P
b
P
Black participants A-Tb Dominant Recessive Additive
1.04 0.10 0.05
0.05 0.54 0.84
0.72 0.05 0.05
0.18 0.75 0.85
White participants A-T Dominant Recessive Additive
0.04 0.04 0.02
0.84 0.89 0.96
0.08 0.12 0.002
0.70 0.66 0.99
a Adjusted for age, systolic blood pressure, pack-year smoking, body mass index, diabetes, low-density lipoprotein cholesterol, coronary heart disease and use of antihypertensive medication and lipidlowering medication (statin). b A-T haplotype of G-6A and M235T.
Journal of Human Hypertension
and A-T haplotype interaction was found in black elderly people (Table 4). Interestingly, there was also a significant haplotype A-T and ACE inhibitor interaction (b ¼ 0.906, P ¼ 0.015 before and b ¼ 1.028, P ¼ 0.013 after adjustment for the selected cardiovascular disease risk factors) for a decreased risk of LEAD in white participants only (Table 5). The use of ACE inhibitors was not associated with ACE_I/D in either white or black participants. The study did not find any other gene– covariate (age, PY-smoking, BMI, diabetes, LDL cholesterol, PCHD, PCEVD, PCHF, and use of statin or other antihypertensive drug) interactions for the risk of LEAD in both black and white participants (data not shown).
Discussion Our study examined the relationship of ACE_I/D variant and G-6A, T174M, and M235T polymorph-
AGT, ACE and LEAD R Li et al 679
Table 4 Gene–gene interaction on the risk of lower extremity arterial disease Adjusteda
Unadjusted b
P
Black participants A-Tb ACE_II A-T*ACE_II
0.008 0.053 0.180
0.96 0.93 0.61
White participants A-T ACE_II A-T*ACE_II
0.182 1.013 0.913
0.24 0.002 0.009
P
b
0.002 0.267 0.110 0.212 1.254 1.069
0.99 0.68 0.77 0.20 0.0006 0.006
Abbreviations: ACE, angiotensin-converting enzyme. a Adjusted for age, systolic blood pressure, pack-year smoking, body mass index, diabetes, low-density lipoprotein cholesterol, coronary heart disease and use of antihypertensive medication and lipidlowering medication (statin). b A-T haplotype of G-6A and M235T was coded as 0, 1, and 2 copies of A-T; A-T*ACE_II refers to the interaction between A-T haplotype and ACE.
Table 5 Gene–drug interaction on the risk of lower extremity arterial disease Unadjusted b
P
Black participants A-Tb ACE_INH A-T*ACE_INH
0.074 0.824 0.298
0.62 0.15 0.38
White participants A-T ACE_INH A-T*ACE_INH
0.164 1.374 0.906
0.28 o0.0001 0.015
Adjusteda b
0.018 0.087 0.040 0.194 1.348 1.028
P
0.91 0.89 0.92 0.23 0.0004 0.013
Abbreviations: ACE, angiotensin-converting enzyme; ACE_INH, ACE inhibitor. a Adjusted for age, systolic blood pressure, pack-year smoking, body mass index, diabetes, low-density lipoprotein cholesterol, coronary heart disease and use of lipid-lowering medication (statin). b A-T haplotype of G-6A and M235T was coded as 0, 1, and 2 copies of A-T; A-T*ACE_INH refers to the interaction between A-T haplotype and ACE inhibitor (ACE_INH: yes or no).
isms of the AGT gene to LEAD in a large population of black and white older adults. The study observed a number of significant differences in the polymorphism frequencies of AGT between the two racial groups. There was evidence that black people possessing the II genotype of ACE or T allele of M235T were more likely to have LEAD. However, after adjustment for multiple testing, the association no longer existed. In whites, there was neither single genetic polymorphism/variant nor haplotype of G6A and M235T associated with LEAD. Interestingly, there were a gene–gene interaction between ACE_II and A-T haplotype of G-6A and M235T and a gene– drug interaction between A-T haplotype and ACE inhibitor for the decreased risk of LEAD.
The statistically significant difference in the distribution of 174M and 235T by race in this study was similar to those reported previously.4,28 Frequencies of G-6A in white participants of this study were similar to those in reports from Paillard et al.29 and Tang et al.30 The D allele frequency of ACE in this study was similar to that in the reports from a meta-analysis of 145 studies,31 except for no significantly racial difference in our study. In addition to the genetic variation, a significant difference in vascular health between black and white elderly people in this study was also found to be consistent with most of the previous studies.32,33 The association between ACE_I/D and cardiovascular diseases is controversial. Some previous studies reported that the D allele of ACE increased the risk of cardiovascular diseases,34,35 but others reported that the ACE I allele was associated with an increased risk of stroke36 and deep vein thrombosis and pulmonary embolism.37 We found that ACE_II was significantly associated with LEAD in black participants only after adjustment for the selected CVD risk factors, but before adjustment for multiple hypotheses testing. After adjustment for multiple hypotheses testing, there was no ACE-LEAD association in black elderly participants, which indicated that the positive ACE-LEAD association in black participants may be due to Type I error. The significant ACE_II and A-T haplotype of AGT interaction and ACE inhibitor and A-T haplotype interaction on the decline of risk for LEAD in white elderly people has not been reported before. The -6A variant at the promoter region of AGT gene increases the basal transcription rate of the AGT gene.7,8 The 235T variant has been consistently associated with higher plasma AGT levels in whites.4,6,11,38,39 Thus, A-T haplotype of G-6A and M235T may increase the AGT level in whites. However, an increased AGT level is not equivalent to an increased level of angiotensin II, a vasoconstrictive agent, because renin cleaves AGT to angiotensin I and ACE converts angiotensin I to angiotensin II. The ACE_I/D variant accounts for 47% of the variance of serum ACE level in whites,10 with the highest serum ACE level in DD carriers and the lowest serum ACE level in II carriers no matter if they were adolescents or centenarians.10,11,40 Thus, people who carry both A-T haplotype of AGT gene and ACE_II genotype may have a low level of the vasoconstrictive agent or have balanced vascular tones, which may decrease the risk of LEAD. The interaction between ACE inhibitors and A-T haplotype for the decreased risk of LEAD further supports the idea that the function of A-T haplotype is suppressed by ACE inhibitor in whites. Thus, it is possible that ACE inhibitor is more effective in the treatment of vascular diseases in whites (43% of whites carrying A-T haplotype) than that in black participants. In fact, this racial difference in the effects of ACE inhibitor for hypertensive treatment was observed in the Antihypertensive and Lipid-Lowering Treatment Journal of Human Hypertension
AGT, ACE and LEAD R Li et al 680
to Prevent Heart Attack Trial (ALLHAT), a randomized, double-blind, active-controlled, clinical outcome trial conducted in approximately 40 000 hypertensive US and Canadian patients aged 55 years or older.41 The ACE gene is located in intron 16, a non-coding site.42 Thus, the ACE activity may be influenced by other polymorphisms within or nearby the ACE gene through the linkage disequilibrium with ACE_I/D. Future studies are needed to investigate other polymorphisms of the ACE gene and their interaction with A-T haplotype for the risk of LEAD in these participants. It should be also interesting to replicate the interaction between ACE inhibitor and A-T haplotype for different cardiovascular outcomes in different age groups of whites. The same interaction was not observed in black participants, which suggested a racial difference in genetic association study. One study of 500 Jamaicans with African origin determined by at least three of four grandparents who were of African origin reported that the mean level of serum ACE was 35% higher in ACE_DD homozygotes compared with that in ACE_II homozygotes (Po0.01), but the genotype was not associated with SBP.43 One other study of serum ACE level and BP with no ACE_I/D genotype data found that serum ACE level was inversely associated with BP in the black participants but positively associated with BP in the whites after adjustment for age, sex, BMI, alcohol consumption, and heart rate.44 This may suggest why the ACE_II genotype and ACE inhibitors did not suppress the function of A-T haplotype in black people. Therefore, A-T haplotype showed an increased risk of LEAD (b ¼ 0.72, P ¼ 0.18) in black participants, although the association did not reach the 0.05 significant level after adjustment for covariates in this study. As this study was conducted in older adults, the issue of selective survival must be considered. It is likely that subjects who were genetically susceptible or environmentally prone to cardiovascular diseases either may not have survived to be included in this study or may not have met the eligibility criteria. For example, the individuals with severe congestive heart failure had to be excluded to meet the criteria of no difficulty walking one-quarter of a mile, climbing 10 steps, or performing basic activities of daily living. Thus, we may be left with a population that is less susceptible to cardiovascular diseases. Indeed, Gardemann et al.45 found, in 2267 white males, an association of the D allele with coronary artery disease in subjects younger than 61.7 years of age but not in subjects aged 61.7 years or older. Another limitation of this study is that the analyses were performed using the baseline crosssectional data. Although chronology is not an issue for gene–phenotype association, because genes are determined before LEAD, it is an issue for drug– disease association. In this report, it is less likely that ACE inhibitors increase the risk of LEAD, but the LEAD patients are more likely to use the Journal of Human Hypertension
medicine. In addition to the chronology, the crosssectional analyses using baseline data did not take into account for the incident LEAD cases in this cohort during follow-up. Thus, the association in this report may be underestimated. In conclusion, this study observed the different distribution of genotypes and haplotypes of AGT genes between black and white elderly people. The homozygous insertion of the ACE gene and ACE inhibitor interacted with A-T haplotype of the AGT gene to decrease the risk of LEAD in whites. The results may provide a clue for future studies of genetic effects on vascular diseases aetiology and pharmacogenetics on vascular disease treatment.
What is known about this topic K Differences in the distribution of G-6A, T174M, and M235T and by race were reported previously.4,28–30 K A significant difference in vascular health between black and white participants was reported in the previous studies.32,33 K The association between angiotensin-converting enzyme (ACE) insertion/deletion polymorphism and cardiovascular diseases is controversial.34,35 What this study adds K We found that ACE_II was significantly associated with lower extremity arterial disease (LEAD) in black participants only after adjustment for the selected cardiovascular disease risk factors, but before adjustment for multiple hypotheses testing. K The significant ACE_II and A-T haplotype of angiotensinogen interaction on the decline of risk for LEAD in white elderly people has not been reported before. K The interaction between ACE inhibitors and A-T haplotype reducing risk of LEAD in white elderly people has not been reported previously.
Disclosures None.
Acknowledgements We acknowledge all Health ABC participants. We appreciate Elizabeth Webb for her editing of this manuscript. This study was supported by grants NIH NIA NO1-AG-6-2101, NO1-AG-6-2103, and NO1-AG-6-2106; R01 AG18702-01A1 and P30 AG021332-01.
References 1 Clauser E, Gaillard I, Wei L, Corvol P. Regulation of angiotensinogen gene. Am J Hypertens 1989; 2: 403–410. 2 Nishimura H, Buikema H, Baltatu O, Ganten D, Urata H. Functional evidence for alternative ANG II-forming pathways in hamster cardiovascular system. Am J Physiol 1998; 275: H1307–H1312.
AGT, ACE and LEAD R Li et al 681
3 Urata H, Nishimura H, Ganten D. Chymase-dependent angiotensin II forming systems in humans. Am J Hypertens 1996; 9: 277–284. 4 Jeunemaitre X, Soubrier F, Kotelevtsev YV, Lifton RP, Williams CS, Charru A et al. Molecular basis of human hypertension: role of angiotensinogen. Cell 1992; 71: 169–180. 5 Jeunemaitre X, Inoue I, Williams C, Charru A, Tichet J, Powers M et al. Haplotypes of angiotensinogen in essential hypertension. Am J Hum Genet 1997; 60: 1448–1460. 6 Schunkert H, Hense HW, Gimenez-Roqueplo AP, Stieber J, Keil U, Riegger GA et al. The angiotensinogen T235 variant and the use of antihypertensive drugs in a population-based cohort. Hypertension 1997; 29: 628–633. 7 Nakajima T, Inoue I, Cheng T, Lalouel JM. Molecular cloning and functional analysis of a factor that binds to the proximal promoter of human angiotensinogen. J Hum Genet 2002; 47: 7–13. 8 Inoue I, Nakajima T, Williams CS, Quackenbush J, Puryear R, Powers M et al. A nucleotide substitution in the promoter of human angiotensinogen is associated with essential hypertension and affects basal transcription in vitro. J Clin Invest 1997; 99: 1786–1797. 9 Schunkert H. Polymorphism of the angiotensin-converting enzyme gene and cardiovascular disease. J Mol Med 1997; 75: 867–875. 10 Rigat B, Hubert C, Alhenc-Gelas F, Cambien F, Corvol P, Soubrier F. An insertion/deletion polymorphism in the angiotensin I-converting enzyme gene accounting for half the variance of serum enzyme levels. J Clin Invest 1990; 86: 1343–1346. 11 Faure-Delanef L, Baudin B, Beneteau-Burnat B, Beaudoin JC, Giboudeau J, Cohen D. Plasma concentration, kinetic constants, and gene polymorphism of angiotensin I-converting enzyme in centenarians. Clin Chem 1998; 44: 2083–2087. 12 Turner ST, Boerwinkle E, Sing CF. Context-dependent associations of the ACE I/D polymorphism with blood pressure. Hypertension 1999; 34: 773–778. 13 Mondorf UF, Russ A, Wiesemann A, Herrero M, Oremek G, Lenz T. Contribution of angiotensin I converting enzyme gene polymorphism and angiotensinogen gene polymorphism to blood pressure regulation in essential hypertension. Am J Hypertens 1998; 11: 174–183. 14 Mondry A, Loh M, Liu P, Zhu AL, Nagel M. Polymorphisms of the insertion/deletion ACE and M235T AGT genes and hypertension: surprising new findings and meta-analysis of data. BMC Nephrol 2005; 6: 1–11. 15 Renner W, Nauck M, Winkelmann BR, Hoffmann MM, Scharnagl H, Mayer V et al. Association of angiotensinogen haplotypes with angiotensinogen levels but not with blood pressure or coronary artery disease: the Ludwigshafen Risk and Cardiovascular Health Study. J Mol Med 2005; 83: 235–239. 16 Keavney B, McKenzie C, Parish S, Palmer A, Clark S, Youngman L et al. Large-scale test of hypothesised associations between the angiotensin-converting-enzyme insertion/deletion polymorphism and myocardial infarction in about 5000 cases and 6000 controls. International Studies of Infarct Survival (ISIS) Collaborators. Lancet 2000; 355: 434–442. 17 Taute BM, Seifert H, Taute R, Glaser C, Podhaisky H. Angiotensin-converting enzyme gene insertion/dele-
18
19
20 21
22 23
24 25 26 27
28
29
30
31
32
33
34
tion polymorphism and carotid artery wall thickness in patients with peripheral arterial occlusive disease. Int Angiol 2000; 19: 337–344. Tseng CH, Tseng CP. Lack of association between angiotensin-converting enzyme gene polymorphism and peripheral vascular disease in type 2 diabetic patients in Taiwan. Circ J 2002; 66: 1014–1018. Renner W, Pabst E, Paulweber B, Malaimare L, Iglseder B, Wascher TC et al. The angiotensin-convertingenzyme insertion/deletion polymorphism is not a risk factor for peripheral arterial disease. Atherosclerosis 2002; 165: 175–178. Newman AB, Sutton-Tyrrell K, Kuller LH. Lowerextremity arterial disease in older hypertensive adults. Arterioscler Thromb 1993; 13: 555–562. Newman AB, Siscovick DS, Manolio TA, Polak J, Fried LP, Borhani NO et al. Ankle-arm index as a marker of atherosclerosis in the Cardiovascular Health Study. Cardiovascular Heart Study (CHS) Collaborative Research Group. Circulation 1993; 88: 837–845. Shanmugam V, Sell KW, Saha BK. Mistyping ACE heterozygotes. PCR Methods Appl 1993; 3: 120–121. Pahor M, Chrischilles EA, Guralnik JM, Brown SL, Wallace RB, Carbonin P. Drug data coding and analysis in epidemiologic studies. Eur J Epidemiol 1994; 10: 405–411. Czika W, Yu X, Wolfinger R. An introduction to genetic data analysis using SAS/Genetics. SAS 2002; 9.0: 1–10. Holland BS, Copenhaver MD. An improved sequentially rejective bonferroni test procedure. Biometrics 1987; 43: 417–424. Holm S. A simple sequentially rejective bonferroni test procedure. Scand J Statist 1979; 6: 65–70. Zaykin DV, Westfall PH, Young SS, Karnoub MA, Wagner MJ, Ehm MG. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered 2002; 53: 79–91. Staessen JA, Kuznetsova T, Wang JG, Emelianov D, Vlietinck R, Fagard R. M235T angiotensinogen gene polymorphism and cardiovascular renal risk. J Hypertens 1999; 17: 9–17. Paillard F, Chansel D, Brand E, Benetos A, Thomas F, Czekalski S et al. Genotype–phenotype relationships for the renin–angiotensin–aldosterone system in a normal population. Hypertension 1999; 34: 423–429. Tang W, Devereux RB, Rao DC, Oberman A, Hopkins PN, Kitzman DW et al. Associations between angiotensinogen gene variants and left ventricular mass and function in the HyperGEN study. Am Heart J 2002; 143: 854–860. Staessen JA, Wang JG, Ginocchio G, Petrov V, Saavedra AP, Soubrier F et al. The deletion/insertion polymorphism of the angiotensin converting enzyme gene and cardiovascular-renal risk. J Hypertens 1997; 15: 1579–1592. Collins TC, Petersen NJ, Suarez-Almazor M, Ashton CM. The prevalence of peripheral arterial disease in a racially diverse population. Arch Intern Med 2003; 163: 1469–1474. Ferreira AV, Viana MC, Mill JG, Asmar RG, Cunha RS. Racial differences in aortic stiffness in normotensive and hypertensive adults. J Hypertens 1999; 17: 631– 637. Sagnella GA, Rothwell MJ, Onipinla AK, Wicks PD, Cook DG, Cappuccio FP. A population study of ethnic variations in the angiotensin-converting enzyme I/D Journal of Human Hypertension
AGT, ACE and LEAD R Li et al 682
35
36
37
38
39
polymorphism: relationships with gender, hypertension and impaired glucose metabolism. J Hypertens 1999; 17: 657–664. O’Donnell CJ, Lindpaintner K, Larson MG, Rao VS, Ordovas JM, Schaefer EJ et al. Evidence for association and genetic linkage of the angiotensin-converting enzyme locus with hypertension and blood pressure in men but not women in the Framingham Heart Study. Circulation 1998; 97: 1766–1772. Bis JC, Psaty BM, Heckbert SR, Lemaitre RN, Smith NL. Angiotensin-converting enzyme insertion/deletion polymorphism, antihypertensive therapy, and risk of nonfatal stroke or myocardial infarction among hypertensive patients. Circulation 2004; 109: 27–28. Wells PS, Rodger MA, Forgie MA, Langlois NJ, Armstrong L, Carson NL et al. The ACE D/D genotype is protective against the development of idiopathic deep vein thrombosis and pulmonary embolism. Thromb Haemost 2003; 90: 829–834. Winkelmann BR, Russ AP, Nauck M, Klein B, Bohm BO, Maier V et al. Angiotensinogen M235T polymorphism is associated with plasma angiotensinogen and cardiovascular disease. Am Heart J 1999; 137: 698–705. Sethi AA, Nordestgaard BG, Tybjaerg-Hansen A. Angiotensinogen gene polymorphism, plasma angiotensinogen, and risk of hypertension and ischemic
Journal of Human Hypertension
40
41
42 43
44
45
heart disease: a meta-analysis. Arterioscler Thromb Vasc Biol 2003; 23: 1269–1275. Bloem LJ, Manatunga AK, Pratt JH. Racial difference in the relationship of an angiotensin I-converting enzyme gene polymorphism to serum angiotensin I-converting enzyme activity. Hypertension 1996; 27: 62–66. Wright Jr JT, Dunn JK, Cutler JA, Davis BR, Cushman WC, Ford CE et al. Outcomes in hypertensive black and nonblack patients treated with chlorthalidone, amlodipine, and lisinopril. JAMA 2005; 293: 1595–1608. Sayed-Tabatabaei FA, Oostra BA, Isaacs A, van Duijn CM, Witteman JC. ACE polymorphisms. Circ Res 2006; 98: 1123–1133. Forrester T, McFarlane-Anderson N, Bennett FI, Wilks R, Cooper R, Rotimi C et al. The angiotensin converting enzyme and blood pressure in Jamaicans. Am J Hypertens 1997; 10: 519–524. He J, Klag MJ, Appel LJ, Charleston J, Whelton PK. The renin–angiotensin system and blood pressure: differences between blacks and whites. Am J Hypertens 1999; 12: 555–562. Gardemann A, Fink M, Stricker J, Nguyen QD, Humme J, Katz N et al. ACE I/D gene polymorphism: presence of the ACE D allele increases the risk of coronary artery disease in younger individuals. Atherosclerosis 1998; 139: 153–159.