Angiotensinogen Gene Variation Associated With Variation in Blood Pressure in Aboriginal Canadians
Abstract We measured blood pressure and related clinical phenotypes in 497 adult native Canadians from an isolated community in Northern Ontario. We analyzed their DNA for genotypes of angiotensinogen. We found that the frequency of the T235 variant of the angiotensinogen gene was 0.89 in this sample. This variant was associated with a significantly increased systolic pressure but not diastolic pressure. We also found that sex and body mass were each highly significantly associated with variation in both systolic and diastolic pressures. We found a significant association between age and variation in systolic pressure but not diastolic pressure. We also found a highly significant association between plasma apolipoprotein B concentration and variation in diastolic pressure but not systolic pressure. The high frequency of the angiotensinogen T235 variant suggests that subjects in this young, essentially normotensive population might be predisposed to hypertension, which may become more apparent in the presence of secondary factors.
Complex quantitative traits such as BP are influenced by genetic and nongenetic factors.1 Candidate genes that determine BP variation include those whose products have a direct role in vascular biology, such as components of the renin-angiotensin system.1 2 Angiotensinogen is a liver protein that interacts with renin to produce angiotensin I, the prohormone of angiotensin II, which increases vascular tone and promotes sodium retention.1 Plasma concentrations of angiotensinogen have been positively correlated with BP, and variation of this intermediate biochemical trait has been associated with variation in the AGT gene.2
One common AGT allele contains a threonine at amino acid residue 235 (T235) and has been associated with increased plasma angiotensinogen concentrations in whites3 4 but not blacks.4 The frequency of the AGT T235 allele ranges from 0.35 in whites to approximately 0.80 in blacks.4 5 The AGT T235 allele has been associated with hypertension in North American,3 French,3 and Japanese6 subjects but not in African,5 Australian,7 or British8 subjects. The AGT T235 allele was also associated with pregnancy-induced hypertension.9 The apparent discrepancies among these various studies may have been related to fundamental genetic differences among the study samples. It is also possible that genomic variation of AGT contributes to a background of genetic susceptibility to hypertension, which only becomes fully expressed in the presence of permissive secondary genetic and/or environmental factors and that these factors differ among different populations.
We postulated that variation in the AGT gene would be associated with variation in BP in a sample of young native (aboriginal) Canadians, who were ascertained through population screening for diabetes. We included as covariates age, body mass, and plasma concentrations of insulin and lipoproteins in addition to genotypes of genes that were not considered a priori to be candidate determinants of BP.
The community of Sandy Lake, Ontario, is located about 2000 km northwest of Toronto, in the subarctic boreal forest region of central Canada. The community is isolated and accessible only by air during most of the year. Historically, the ancestors of the contemporary residents of this region lived a nomadic, hunting-gathering subsistence typical of other Algonkian-speaking peoples of the northeastern subarctic. Since the development of the reservation and residential school systems, the lifestyle of these people changed from very physically active to very sedentary. The primary source of food changed from wildlife with roots and berries to processed foods high in animal fats, which are supplied by a company store.
A total of 728 members of this community aged 10 years and older participated in the Sandy Lake Health and Diabetes Project, an initiative undertaken to determine the prevalence of non–insulin-dependent diabetes mellitus in this population. Assessments included a questionnaire for medical history, including the current use of antihypertensive medications. Physical examination included determination of BMI defined as weight (kilograms) divided by height (meters) squared and two separate BP determinations in the right arm with the volunteer seated. Systolic BP was recorded to the nearest 2 mm Hg at the appearance of the first Korotkoff sound (phase I), and diastolic BP was recorded to the nearest 2 mm Hg at the disappearance of the fifth Korotkoff sound (phase V). Plasma samples were obtained with informed consent after subjects had fasted 10 to 12 hours. Exclusion criteria included age younger than 18 years and an inadequate blood sample for all determinations. The project was approved by The University of Toronto Ethics Review Committee.
Biochemical and Genetic Analyses
Sufficient DNA and phenotypic information were obtained for analysis from 497 subjects aged between 18 and 74 years. Of these, 57.0% were women. The means±SD for age, systolic BP, diastolic BP, and BMI were, respectively, 35.0±13.5 years, 119.4±15.0 mm Hg, 68.3±11.4 mm Hg, and 28.1±5.26 kg/m2. Blood for plasma insulin determination was centrifuged at 3000 rpm for 10 minutes, and the plasma was stored at −70°C. Fasting plasma insulin concentrations were determined by radioimmunoassay (Pharmacia). Blood for lipoprotein analyses was centrifuged at 2000 rpm for 30 minutes, and the plasma was stored at −70°C. Fasting plasma concentrations of triglycerides and total, LDL, and high-density lipoprotein cholesterols as well as apoA-I and apoB were determined as described.10 11 12 13 Established procedures were used for determination of genotypes of AGT codons 174 and 235,10 intestinal fatty acid binding protein (FABP2) codon 54,14 apoE (APOE) exon 4,15 and arginine vasopressin receptor (AVPR2) codon 309.16 Known genotypic standards were included for each electrophoresis.
SAS (version 6.1) was used for all statistical comparisons.17 The distribution of systolic and diastolic BPs was significantly non-normal in this data set. Therefore, for parametric statistical analyses, each quantitative variable was transformed and subjected to analysis of normality as described.10 11 12 13 The transformed variables were used for parametric statistical analyses, but the nontransformed values are presented in the tables.
ANOVA was performed using the general linear models procedure for determination of the sources of variation for systolic and diastolic BPs, with F tests computed from the type III sums of squares.17 This form of sums of squares is applicable to unbalanced study designs. Dependent variables were transformed systolic BP and diastolic BP. Independent variables were age, sex, the natural logarithm of BMI, and current treatment with an antihypertensive medication. Also included as an independent variable were the natural logarithms of the plasma concentrations of insulin and apoB. Finally, all genotypes were included as independent variables.
When a significant association between a genetic variable and BP was identified with the ANOVA, BP differences between individuals classified by genotype were compared using a t test, with Bonferroni correction for multiple comparisons.17 For significant associations between a continuous variable and BP identified by ANOVA, the Pearson correlation coefficient was calculated.17
Allele and Genotype Frequencies
Allele frequencies are shown in Table 1⇓. Genotype frequencies did not deviate from those predicted by the Hardy-Weinberg law in this study sample (data not shown). The observed frequency of the AGT M174 allele was more than twice that reported in whites.3 10 The observed frequency of the AGT T235 allele was higher than that reported in white,3 10 black,4 5 and Japanese6 populations. The frequency of the FABP2 T54 allele was approximately one half that reported in Pima Indians.14 The frequency of the APOE E2 allele was among the lowest reported in any human population.18 The frequency of the AVPR2 allele with adenosine at the third nucleotide of codon 309 was 0.19, which was approximately one quarter the frequency reported in Scandinavians.16
Genetic Determinants of Variation in Systolic and Diastolic BPs
The results of the ANOVA are shown in Table 2⇓. One ANOVA was performed each for systolic and diastolic BPs. Since ANOVA takes multiple comparisons into account, we did not have to adjust the levels of nominal significance.
Systolic BP was significantly associated with AGT codon 235 genotype (P=.037) but not with any other genetic or biochemical variable. Diastolic BP was not significantly associated with any genomic variation.
Homozygotes for the AGT M235 allele and heterozygotes had the lowest mean systolic BP (115.0±18.5 and 115.4±12.6 mm Hg, respectively), whereas homozygotes for the AGT T235 allele had the highest mean systolic BP (120.3±15.2 mm Hg). Pairwise comparisons showed that mean systolic BP in homozygotes for the AGT T235 allele was significantly higher than in heterozygotes (P<.05, Bonferroni T test).
In this sample, 36 subjects (7.2%) had a systolic BP exceeding 140 mm Hg and/or a diastolic BP exceeding 90 mm Hg. The frequency of the AGT T235 allele in this hypertensive subset was 0.91, which was not significantly different from the prevalence of 0.89 seen in the nonhypertensive subset. Also, 40 subjects (7.9%) were prescribed antihypertensive medications. The frequency of the AGT T235 allele in those taking antihypertensive medications was 0.94, which was significantly different from the prevalence of 0.89 seen in those not taking antihypertensive medications (χ2=5.9, P=.015).
Other Factors Associated With Variation in BP
Significant associations were identified between transformed systolic BP and the independent variables of sex, log BMI, and current treatment with antihypertensive medications (all P<.0001). After adjustment for other variables, systolic BP was higher in men than women (122.3±14.7 versus 117.8±15.6 mm Hg, P<.0001). In contrast to diastolic BP, systolic BP was very significantly associated with age (P<.0001). The Pearson correlation coefficient between systolic BP and age was 0.43 (P<.0001). In contrast to diastolic BP, no plasma biochemical trait was associated with variation in systolic BP (data not shown).
Significant associations were identified between transformed diastolic BP and the independent variables of sex, log BMI, and current treatment with antihypertensive medications (P=.0015, P=.0009, and P=.0009, respectively). After adjustment for other variables, diastolic BP was higher in men than women (70.2±12.2 versus 66.6±10.1 mm Hg, P<.0001). In contrast to systolic BP, diastolic BP was very significantly associated with plasma apoB concentrations (P<.0001). The Pearson correlation coefficient between diastolic BP and plasma apoB was 0.39 (P<.0001). A similar association and correlation were found when plasma LDL cholesterol concentrations were substituted for apoB (data not shown). However, no other lipoprotein or apolipoprotein trait was associated with variation in diastolic BP (data not shown).
The principal findings from this study of aboriginal Canadians are (1) a significant association between variation in systolic BP and variation in the AGT gene at codon 235, (2) a highly significant association between age and variation in systolic BP but not diastolic BP, (3) a highly significant association between plasma apoB concentration and variation in diastolic BP but not systolic BP, and (4) allele frequencies of all genes studied in this population that were unique among human populations, particularly the extremely high frequency of AGT T235.
Despite a frequency of AGT T235 of 0.89, only 40 members of this study group (7.9%) were prescribed antihypertensive medications. The frequency of the AGT T235 allele in the subset who were taking antihypertensive medications was 0.94, which was significantly different from the prevalence of 0.89 seen in the subset who were not taking antihypertensive medications. Also, only 36 members of this study group (7.2%) had a systolic BP exceeding 140 mm Hg and/or diastolic BP exceeding 90 mm Hg. The frequency of the AGT T235 allele in this hypertensive subset was 0.91, which was not different from the prevalence of 0.89 in the normotensive subjects. The low incidence of hypertension in this aboriginal study sample despite the very high frequency of the AGT T235 allele may have been related to the relatively young age of the study sample. Alternatively, there may be unique secondary factors in the genetic background of this study sample that might attenuate the development of hypertension.
Jeunemaitre et al3 found that variation of AGT codon 235 was strongly associated with severe hypertension. They demonstrated that the AGT T235 variant was significantly more frequent in severely hypertensive index cases compared with normotensive control subjects.3 This was subsequently confirmed in Japanese subjects6 but not in white subjects living in Australia7 or England.8 The reasons for these discrepancies may relate to fundamental genetic differences in the study samples, within which subjects were ascertained with the use of different criteria. The strong association of AGT T235 with hypertension in the Paris and Salt Lake City samples suggests that it may be a factor predisposing to hypertension in carriers.3 The results in our aboriginal sample suggest that it might have a subtle, preclinical effect upon resting BP. Furthermore, the expression of phenotypic hypertension in AGT T235 carriers probably requires the presence of secondary factors.
We also observed in our aboriginal study sample a significant positive correlation between age and systolic BP but not diastolic BP. We observed a similar age-specific association with systolic BP in North American Hutterites,19 suggesting that this may be a more general association. These observations suggest that aging-related factors, such as arterial vessel wall stiffness or related physical properties, may be more important determinants of systolic BP than diastolic BP.
We also observed in our aboriginal study sample a significant positive correlation between plasma apoB concentration and diastolic BP but not systolic BP. We have observed a similar apoB-specific association only with diastolic BP in North American Hutterites,19 suggesting that this too may be a more general association. These observations suggest that factors related to apoB-containing lipoprotein metabolism may be more related to diastolic BP than systolic BP.
ApoB is the sole protein component of LDL, whose plasma levels are associated with an increased risk of atherosclerosis.20 In the Sandy Lake sample, we found a concentration-dependent positive correlation between plasma apoB-containing lipoproteins and diastolic BP but not systolic BP. Several epidemiological studies have reported a general association of lipid abnormalities with essential hypertension.21 22 23 Plasma apoB and diastolic BP may share common determinants. For example, a fundamental defect in glucose disposal has been proposed to underlie the syndrome that includes compensatory hyperinsulinemia, elevated plasma apoB-containing lipoproteins, and hypertension.24 Hyperinsulinemia has been proposed to directly raise both apoB and BP.24 However, plasma concentrations of insulin were not associated with either systolic or diastolic BP in a multivariate analysis (Table 2⇑) or in univariate analyses (data not shown). This suggests that insulin did not mediate the association between plasma apoB concentration and diastolic BP.
It is possible that the association between diastolic BP and plasma apoB concentrations may have been related to an effect of apoB on vascular tone. For example, oxidation of apoB-containing lipoproteins, particularly LDL, impairs endothelium-mediated relaxation in arterial segments.25 This suggests that changes in vascular responsiveness that could predispose to hypertension might be secondary to changes in plasma lipoprotein concentrations. Furthermore, lowering of apoB has a beneficial effect on endothelium-mediated responsiveness of the coronary arteries.26 27 Recent in vitro studies suggest that angiotensin II can cause a nonoxidative modification of apoB-containing lipoproteins, thereby enhancing their uptake by cellular components of the vascular wall.28
Finally, we observed no association between genotypes of FABP2 codon 54, APOE exon 4, or AVPR2 codon 309 with either systolic or diastolic BP. These variables were included as genetic controls; there was no mechanistic reason a priori to suspect that they might regulate BP. Remarkably, the minor allele frequencies of these genes were quite different in the Sandy Lake population than in most other nonaboriginal populations reported. These differences could be due either to the fundamental genetic uniqueness of this Canadian aboriginal population or to more recent founder effects specifically involving the ancestors of the contemporary community.
In summary, we have observed that genomic variation of AGT codon 235, in addition to age, sex, and body mass, was associated with interindividual variation in systolic BP in a young, Canadian aboriginal population. Plasma concentrations of apoB, sex, and BMI were each associated with variation in diastolic BP. There is a very high frequency in this population of the AGT T235 allele, which has been associated with hypertension in other populations. We now have the opportunity to follow this community prospectively for the development of hypertension with advancing age and Westernization.
Selected Abbreviations and Acronyms
|BMI||=||body mass index|
This work was supported by grants from the National Institutes of Health (91-DK-01), the Ontario Ministry of Health (No. 04307), the Heart and Stroke Foundation of Ontario (No. B3073), the Medical Research Council of Canada (No. MA-13430), and the St Michael’s Hospital Foundation. Dr Hegele is a Career Investigator (No. CI-2979) of the Heart and Stroke Foundation of Ontario. The authors would like to acknowledge the following groups and individuals, whose cooperation was essential in the design and implementation of this project: the chief and council of the community of Sandy Lake, the Sandy Lake community surveyors, the Sandy Lake nurses, the staff of the University of Toronto Sioux Lookout program, Annette Barnie, Martin Jugenberg (AGT codon 174 genotypes), Jon Eperjesi (AGT codon 235 genotypes), Stefan Sadikian (FABP2 genotypes), Ulana Kawun (AVPR2 genotypes), Teresa Lippingwell Liling Chan, and the Department of Clinical Epidemiology of the Samuel Lunenfeld Research Institute.
- Received July 24, 1996.
- Revision received October 21, 1996.
- Accepted November 4, 1996.
Ward RH. Familial aggregation and genetic epidemiology of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis and Management. New York, NY: Raven Press Publishers; 1990:81-100.
Williams RR, Hunt SC, Hopkins PN, Hasstedt SJ, Wu LL, Lalouel JM. Finding the genes for human hypertension. In: Woodford FP, Davignon J, Sniderman A, eds. Atherosclerosis X. Amsterdam, Netherlands: Elsevier Science BV; 1994:856-861.
Bloem LJ, Manatunga AK, Tewksbury DA, Pratt JH. The serum angiotensinogen concentration and variants of the angiotensinogen gene in white and black children. J Clin Invest. 1995;95:948-953.
Rotimi C, Morrison L, Cooper A, Oyejide C, Effing E, Ladipo M, Osotemheninen O, Ward R. Angiotensinogen gene in human hypertension: lack of an association of the T235 allele among African Americans. Hypertension. 1994;24:591-594.
Hata A, Namikawa C, Sasaki M, Sato K, Nakamura T, Tamura K, Lalouel J-M. Angiotensinogen as a risk factor for essential hypertension in Japan. J Clin Invest. 1994;93:1285-1287.
Hegele RA, Brunt JH, Connelly PW. A polymorphism of the angiotensinogen gene is associated with variation in blood pressure in a genetic isolate. Circulation. 1994;90:2207-2212.
Hegele RA, Evans AJ, Tu L, Ip G, Brunt JH, Connelly PW. A gene-gender interaction affecting lipoproteins in a genetic isolate. Arterioscler Thromb. 1994;14:671-678.
Hegele RA, Brunt JH, Connelly PW. A polymorphism of the paraoxonase gene associated with variation in blood pressure in a genetic isolate. Arterioscler Thromb Vasc Biol. 1995;15:89-95.
Hegele RA, Brunt JH, Connelly PW. Multiple genetic determinants of variation of plasma lipoproteins in a genetic isolate. Arterioscler Thromb Vasc Biol. 1995;15:861-871.
Baier LJ, Sacchettini JC, Knowler WC, Eads J, Paolisso G, Tataranni PA, Mochizuki H, Bennett PH, Bogardus C, Prochazka M. An amino acid substitution in the human intestinal fatty acid binding protein is associated with increased fatty acid binding, increased fat oxidation and insulin resistance. J Clin Invest. 1995;95:1281-1287.
Hixson JE, Vernier DT. Restriction isotyping of the human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res. 1990;31:545-548.
Friedman E, Carson E, Larsson C, DeMarco L. A polymorphism in the coding region of the vasopressin type 2 receptor (AVPR2) gene. Hum Mol Genet. 1993;2:1746.
SAS Institute Inc. SAS/STAT Guide for Personal Computers, Version 6 Edition. Cary, NC: SAS Institute Inc; 1987.
Hegele RA, Brunt JH, Connelly PW. Genetic and biochemical factors associated with variation in blood pressure in a genetic isolate. Hypertension. 1996;27:308-312.
Stamler J, Wentworth D, Neaton JD. Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356 222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 1986;256:2823-2828.
Reaven GM. Role of insulin resistance in human disease. Diabetes. 1988;37:1595-1607.
Simon BC, Cunningham LD, Cohen RA. Oxidized low density lipoproteins cause contraction and inhibit endothelium-dependent relaxation in the pig coronary artery. J Clin Invest. 1990;86:75-79.
Treasure CB, Klein JL, Weintraub WS, Talley JD, Stillabower ME, Kosinski AS, Zhang J, Boccuzzi SJ, Cedarholm JC, Alexander RW. Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease. N Engl J Med. 1995;332:481-487.
Keidar S, Kaplan M, Aviram M. Angiotensin II-modified LDL is taken up by macrophages via the scavenger receptor, leading to cellular cholesterol accumulation. Arterioscler Thromb Vasc Biol. 1996;16:97-105.