Blood Pressure and the M235T Polymorphism of the Angiotensinogen Gene
The angiotensinogen gene locus (1q42-43) has been linked to hypertension in affected relative-pair studies (including a previous UK study), but the role of the Met→Thr polymorphism at position 235 remains controversial. Using this marker, we investigated the relationship between angiotensinogen genotype and blood pressure in two data sets from the East Anglia region of the United Kingdom. Two hundred twenty-three untreated hypertensive and 187 normotensive control subjects were recruited through local general practices. Blood pressure (including pretreatment measurements in the hypertensive group), age, sex, body mass index, alcohol consumption, cholesterol level, and angiotensinogen genotype were recorded for all subjects. The influence of angiotensinogen genotype on blood pressure was assessed with a general linear model ANOVA with adjustment for age, sex, body mass index, and alcohol consumption. There was no evidence for an association between angiotensinogen genotype and blood pressure level in either the hypertensive or normotensive data set. Angiotensinogen genotype did not influence blood pressure in subjects aged <50 years, women, or those with a body mass index <26 kg/m2. We conclude that the angiotensinogen Met→Thr polymorphism is not a marker for blood pressure level in these East Anglian subjects. Further studies are required to confirm the involvement of the 1q locus in the development of hypertension in UK subjects and to delineate the functional variant(s) in this chromosomal region that influences blood pressure.
Several lines of evidence point to the involvement of the AGT gene in BP regulation. Physiological studies have shown that the renin-angiotensin system plays a major role in salt/water homeostasis and vascular tone regulation.1 Epidemiological surveys have shown a correlation between plasma AGT and BP levels2 and higher BP and plasma AGT levels in the offspring of hypertensives.3 Mice in which the AGT gene has been deleted by homologous recombination are hypotensive,4 whereas in mice with duplications of the AGT gene, BP and plasma AGT levels are positively correlated with the number of gene copies.5 In addition, mice rendered doubly transgenic for human renin and AGT genes are hypertensive.6 In contrast, specific AGT antisense oligonucleotides transiently depress BP when administered into the central nervous system of spontaneously hypertensive rats.7
In humans, the AGT locus has been linked to essential hypertension in American, French, British, and African Caribbean pedigrees.8 9 10 A number of polymorphic variants of the mature AGT peptide have also been identified,8 but with the exception of one rare variant (L10F) identified in a single individual with preeclampsia,11 none has yet been shown to alter the kinetics of the interaction between AGT and renin. However, a variant that results from a Met→Thr substitution at codon 235 (M235T) has been shown to track weakly with plasma AGT levels,8 12 suggesting that this polymorphism may be a marker in linkage disequilibrium with a variant that regulates AGT gene transcription or mRNA stability. Association (case-control) studies of the M235T polymorphism in essential hypertension have yielded conflicting results,13 14 but the majority have examined hypertension and normotension as simple categorical variables. Three studies have examined the relation between quantitative variations in BP and AGT polymorphism: One was conducted in a genetic isolate in the United States (which is unlikely to be representative of the wider US population),15 and the other two were conducted in Japan,16 17 where allele frequencies of the M235T polymorphism differ from those reported for western European white subjects. We investigated the relationship between the AGT M235T polymorphism and BP level in two groups: (1) a population-based sample of healthy subjects identified through local family physicians in the East Anglia region of the United Kingdom and (2) a group of hypertensive patients referred to a specialist hypertension clinic located within the same geographic catchment area. Baseline BP recordings were made before antihypertensive therapy was initiated, thereby allowing a quantitative analysis of the contribution of the M235T variant to BP variation in both groups.
To reduce the effects of population stratification, DNA was collected from two entire cohorts of randomly selected local hypertensive subjects for whom pretreatment BPs were available and from a subset of normotensive subjects from a local population screening who most closely matched the hypertensive subjects in terms of age, BMI, and sex.
Hypertensive patients were drawn from the database of the HRC at Addenbrooke's Hospital in Cambridge, which serves a local population of ≈400 000 individuals. In 1986 contact was made with local general practitioners, who were encouraged to refer all new patients with elevated BP to the HRC before antihypertensive therapy was initiated. At the HRC secondary causes of hypertension were excluded by routine investigations, and patients who fulfilled the entry criterion of having a BP >160/90 mm Hg on three occasions over 3 months were randomized by computer to receive one of four classes of antihypertensive drug monotherapy for 4 weeks. Baseline pretreatment BP, as measured in the recumbent position by a Datascope Accutor 2000 (Datascope Corp), and BMI were recorded on the day of randomization, and a blood sample was collected for DNA extraction and random cholesterol measurement. A target BP of <160/90 mm Hg was achieved with appropriate changes in drug dosage or replacement of drug monotherapy with combination therapy after the initial 4-week treatment phase. Thereafter, patients were followed up in hospital every 2 years, with BP monitoring by their general practitioners in the intervening period. To date, 663 patients have been recruited; of these, 223 who were randomized to ACE inhibitor or β-blocker therapy and for whom DNA data were available are reported herein. In part, these subjects were chosen to allow an additional investigation into the effect of renin-angiotensin gene polymorphisms on the BP response to ACE inhibitor treatment (the β-blocker–treated group served as a control). Since treatment allocation was random, these subjects are representative of all hypertensive cases seen in the HRC (data not shown). All subjects were white. A portion of the data on the ACE inhibitor–treated subjects and selection criteria have been reported previously.18
Control subjects were drawn from a large community survey of cardiovascular risk factors conducted in the same region from broadly the same general practitioners who had referred the hypertensive cases to the HRC. Between 1990 and 1994, 32 000 subjects aged 40 to 69 years with no history of cardiovascular or other disease were identified from general practice registers and invited to outreach clinics at 43 local general practices for health screening. Subjects were questioned about diet, smoking habits, alcohol consumption, and family history of illness. Casual BP (mean of two readings) was measured in the sitting position with the Datascope 2000 automated device and BMI was recorded. A random cholesterol level was measured from a sample of fingerstick blood on a Reflotron analyzer (Boehringer Mannheim UK). Blood for DNA analysis was also collected from control subjects at four of these general practices. One hundred eighty-seven of these subjects (all white) who were most closely matched to the hypertensive subjects in terms of age, sex, and BMI were included in the present study. Our own separate analysis (data not shown) has indicated that these control subjects are representative of the entire normotensive group identified by this screening.
DNA was extracted from peripheral blood leukocytes by using a standard protocol.19 The AGT M235T polymorphism was typed by the mismatch priming method of Russ et al20 with some modifications. In brief, 100 ng of genomic DNA was amplified for 35 cycles in a reaction volume of 25 μL containing 0.2 mmol/L of each dNTP, 10 mmol/L Tris HCl (pH 9.0), 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.1% Triton X-100, 1 U Taq DNA polymerase, and 25 pmol each of primers 5′-GATGCGCACAAGGTCCTGTC-3′ (forward) and 5′-CAGGGTGCTGTCCACACTGGACCCC-3′ (reverse). The 303-bp polymerase chain reaction product was exposed to the restriction enzyme TthIII I for 16 hours at 65°C and electrophoresed on a 3% agarose gel with ethidium bromide staining. With this primer pair, the M235 allele is visible as an uncut 303-bp fragment and the T235 allele as a cleaved 279-bp fragment.
Statistical analysis was performed with minitab software (Minitab Inc). Genotype and allele frequencies in control and hypertensive groups were compared by χ2 analysis. Continuous variables were compared between hypertensive and control groups by Student's t test (or the Mann-Whitney U test for nonnormally distributed variables). The influence of AGT genotype on continuous variables was investigated by one-way ANOVA (or the Kruskal-Wallis test for nonnormally distributed variables). In addition, the effect of AGT genotype on BP was investigated with the general linear model ANOVA with adjustment for age, sex, BMI, and alcohol consumption. Multiple regression analysis was also performed with SBP or DBP as the dependent variable and age, BMI, alcohol consumption, and AGT genotype (coded 0, 1, or 2 according to the number of T235 alleles) as independent variables.
The characteristics of the hypertensive and control populations are shown in Table 1⇓. Although normotensive subjects were younger, there was no significant difference between the two groups with respect to BMI, alcohol consumption, or cholesterol level. BPs displayed a continuous unimodal distribution in both data sets and were significantly lower in control subjects.
Genotype and Allele Frequencies
Genotype and allele frequencies in the control and hypertensive groups were comparable with those in other western European white subjects. There was no detectable distortion from Hardy-Weinberg equilibrium in either data set (χ2=0.038, df=2, P>.5 for cases and χ2=0.057, df=2, P>.5 for control subjects), and there was no significant difference in genotype or allele frequencies between hypertensive and normotensive groups (Table 2⇓).
Genotype and BP Variation
Table 3⇓ shows BP and other variables according to AGT genotype in hypertensive and normotensive data sets. No significant difference was observed between genotype classes for unadjusted or adjusted BP, nor for any other measured variable. By multiple regression analysis, age (P=.0001 for SBP and P=.006 for DBP) and sex (P<.021 for SBP and P<.0001 for DBP) but not AGT genotype were predictors of BP in the control group (R2=.155, P<.0001 for SBP; R2=.134, P<.0001 for DBP). Only age (P<.0001 for both SBP and DBP) was a predictor of BP variation in the hypertensive group (R2=.176, P<.0001 for SBP; R2=.071, P<.008 for DBP). Because previous studies reported relationships between AGT polymorphism and hypertension to be enhanced in women, younger subjects, or those with a lower BMI, ANOVA was repeated for women alone, subjects aged <50 years, and subjects with a BMI <26 kg/m2. No significant influence of AGT genotype on BP was observed within any of these subsets in either the hypertensive or control group (data not shown).
In this study conducted in the East Anglia region of the United Kingdom, we found no evidence of an association between the M235T polymorphism of the AGT gene and BP in either hypertensive or normotensive subjects. Our findings contrast with those of Hata et al13 and Jeunemaitre et al8 but are consistent with other reports about this particular polymorphism.9 14
Three separate studies of hypertensive sibling pairs have demonstrated linkage between essential hypertension and a dinucleotide repeat polymorphism downstream from the AGT gene on chromosome 1q.8 9 10 Linkage studies that use such an approach examine a relatively small number of meioses and consequently implicate a relatively large chromosomal region within which functional variants that determine susceptibility to hypertension presumably lie. Positive findings from linkage analysis led to the hypothesis that the AGT gene itself might contain such functional variants. Molecular scanning of the AGT gene by SSCP analysis has identified 15 polymorphisms, including the M235T variant studied herein.8 Since none of the eight biallelic polymorphisms located in the coding region of the gene affect critical regions of the AGT molecule, it is thought unlikely that they alter the enzymology of the AGT interaction with renin (the rate-limiting step in the renin-angiotensin cascade). However, the finding that homozygosity for the T235 allele is associated with higher concentrations of circulating AGT8 12 invoked a mechanism whereby AGT gene variants might elevate BP and led to speculation that this variant might be in linkage disequilibrium with a polymorphism that regulates the rate of AGT gene transcription or mRNA stability. Although this hypothesis has not been tested formally, case-control studies have shown an association of the T235 allele with hypertension in some13 14 but not all populations. The T235 variant has also been associated with the development of preeclampsia21 and coronary artery disease22 23 in case-control studies, but again, such positive findings are not universal.24 25
How is it possible to reconcile the discrepancies both within and between association (case-control) and linkage (affected sibling-pair) studies? Case-control studies are sensitive to the effects of undetected selection bias, population stratification, confounding by other variables, and the definition of hypertension used. The hypertensive subjects in the present study are likely to be more representative of the wider population of hypertensives normally managed by family physicians. Such hypertensives contrast with more severely affected subjects, who are normally seen in hospital tertiary referral clinics, are usually already being treated, may have secondary causes for their hypertension, and are often on multidrug therapy. We would therefore argue that the risk of population stratification in our study is lower than in most others. Furthermore, by relating AGT M235T genotype to individual BP in a large sample of individuals, we hypothesized that we could overcome many of the difficulties that normally accompany population association studies. The recording of pretreatment BPs under carefully controlled conditions in a group of “never-treated” hypertensives and a group of healthy control subjects from the same geographic region allowed us to explore this relationship in two data sets. However, we found no association between genotype and BP in either the hypertensive or the normotensive group despite adjustment for possible confounding variables. Neither was any association detected in subgroups of younger subjects, women, or those with a BMI <26 kg/m2. In the hypertensive data set (in which BP variance was greater), we had 90% power to detect an 8 mm Hg BP difference between MM and TT homozygotes at P=.05. Thus, it is possible that a smaller effect of the M235T polymorphism may have gone undetected. However, it is instructive to recall that the M235T polymorphism also failed to associate with hypertension in a case-control study in the southeast United Kingdom, despite evidence for linkage of the AGT locus to hypertension in affected sibling pairs, half of whom formed the “hypertensive” arm of the case-control study.9
Population association studies are dependent on the phenomenon of linkage disequilibrium, which occurs when two polymorphisms (eg, marker and functional variant) are in physical proximity on the same chromosome such that the likelihood of genetic recombination between the two alleles at meiosis is small. For linkage disequilibrium to be maintained in a population of unrelated subjects after many millions of meioses, the “disease allele” must have arisen from a small number of common ancestors and the rate of new mutation at either marker or disease locus must be low. Since conclusions drawn from observations based on linkage disequilibrium can be applied only over relatively small genetic distances, our findings do not exclude the whole AGT locus from involvement in BP regulation. Nevertheless, taken together with the negative findings of the previous association study in the southeastern United Kingdom, the present investigation suggests that a major BP-determining variant at the AGT locus has yet to be defined for the UK population. The T235 variant may still be a marker for more narrowly defined phenotypes relating to the renin-angiotensin cascade. Hopkins et al26 have demonstrated blunted renal blood flow responses to angiotensin II infusion in subjects with the TT genotype. This is one characteristic that distinguishes a subset of essential hypertensives who have been classified as “nonmodulators,”27 an additional feature being their enhanced sensitivity to ACE inhibition. In support of this observation, we previously identified an association between the T235 allele and BP response to ACE inhibitor therapy.18
Despite SSCP analysis of the five exons and promoter region of the AGT gene in American/French populations, wherein both the association and the linkage to this locus have been demonstrated,8 the hypertension allele or haplotype has not been found. This suggests that the susceptibility mutation may lie some distance from the AGT gene itself, although failure of the SSCP method (which has a sensitivity of 80%) to detect a functional mutation in this region cannot be excluded.
We conclude that the T235 variant of the AGT gene is not a marker for a BP susceptibility allele at this locus in East Anglian subjects, but as with any negative finding from an association study, the AGT gene cannot be completely excluded as a BP-determining gene in our population. Our findings emphasize the difficulties that are likely to be encountered in discovering other causative mutations for complex genetic disorders, such as hypertension.
Selected Abbreviations and Acronyms
|BMI||=||body mass index|
|HRC||=||Hypertension Research Clinic|
|(S/D)BP||=||(systolic/diastolic) blood pressure|
|SSCP||=||single-strand conformational polymorphism|
This study was supported by a grant from the British Heart Foundation. A.D.H. is a Medical Research Council Clinical Training Fellow and holds a Raymond and Beverley Sackler studentship. P.S. is a British Heart Foundation Junior Research Fellow.
Reprint requests to A.D. Hingorani, Clinical Pharmacology Unit, University of Cambridge, Level 2, F&G Block, Addenbrooke's Hospital, Cambridge, CB2 2QQ, UK.
- Received June 22, 1996.
- Revision received July 11, 1996.
- Accepted August 23, 1996.
Bader M, Paul M, Fernandez-Alfonso M, Kaling M, Ganten D. Molecular biology and biochemistry of the renin-angiotensin system. In: Swales JD, ed. Textbook of Hypertension. 1st ed. Oxford, England: Blackwell Scientific Publications; 1994:214-232.
Walker WG, Whelton PK, Saito H, Russel RP, Hermann J. Relation between blood pressure and renin, renin substrate angiotensin II, aldosterone and urinary sodium and potassium in 574 ambulatory subjects. Hypertension. 1979;1:287-291.
Watt GCM, Harrap SB, Foy CJW, Holton DW, Edwards HV, Davidson HR, Connor JM, Lever AF, Fraser R. Abnormalities of glucocorticoid metabolism and the renin-angiotensin system: a four-corners approach to the identification of genetic determinants of blood pressure. J Hypertens. 1992;10:473-482.
Tanimoto K, Sugiyama F, Goto Y, Ishida J, Takimoto E, Yagami KI, Fukamizu A, Murakami K. Angiotensinogen-deficient mice with hypotension. J Biol Chem. 1994;269:31334-31337.
Kim HS, Krege JH, Kluckman KD, Hagaman JR, Hodgin JB, Best CF, Jennette JC, Coffman TM, Maeda N, Smithies O. Genetic control of blood pressure and the angiotensinogen locus. Proc Natl Acad Sci U S A. 1995;92:2735-2739.
Fukamizu A, Sugimura K, Takimoto E, Sugiyama F, Seo MS, Takahashi S, Hatae T, Kajiwara N, Yagami KI, Murakami K. Chimeric renin-angiotensin system demonstrates sustained increase in blood pressure of transgenic mice carrying both human renin and human angiotensinogen genes. J Biol Chem. 1993;268:11617-11621.
Jeunemaitre X, Soubrier F, Kotelevtsev YV, Lifton RP, Williams CS, Charru A, Hunt SC, Hopkins PN, Williams RR, Lalouel J-M, Corvol P. Molecular basis of hypertension: role of angiotensinogen. Cell. 1993;71:169-180.
Caulfield M, Lavender P, Newell-Price J, Farrall M, Kamdar S, Daniel H, Lawson M, De Freitas P, Fogarty P, Clark AJL. Linkage of the angiotensinogen gene locus to human essential hypertension in African Caribbeans. J Clin Invest. 1995;96:687-692.
Inoue I, Rohrwasser A, Helin C, Jeunemaitre X, Crain P, Bohlender J, Lifton RP, Corvol P, Ward K, Lalouel JM. A mutation of angiotensinogen in a patient with preeclampsia leads to altered kinetics of the renin-angiotensin system. J Biol Chem. 1995;270:11430-11436.
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.
Hata A, Namikawa C, Sasaki M, Sato K, Nakamura T, Tamura K. 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 associated with variation in blood pressure in a genetic isolate. Circulation. 1994;90:2207-2212.
Iwai N, Shimoike H, Ohmichi N, Kinoshita M. Angiotensinogen gene and blood pressure in the Japanese population. Hypertension. 1995;25(pt II):688-693.
Hingorani AD, Jia H, Stevens PA, Hopper R, Dickerson JEC, Brown MJ. Renin-angiotensin system gene polymorphisms influence blood pressure and the response to angiotensin converting enzyme inhibition. J Hypertens. 1995;13(pt II):1602-1609.
Lahiri DK, Nurnberger JI. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res. 1991;19:5444.
Russ AP, Maerz W, Ruzicka V, Stein U, Gross W. Rapid detection of the hypertension associated Met235 Thr allele of the human angiotensinogen gene. Hum Mol Genet. 1993;2:609-610.
Ishigami T, Umemura S, Iwamoto T, Tamura K, Hibi K, Yamaguchi S, Nyuui N, Kimura K, Miyazaki N, Ishii M. Molecular variant of angiotensinogen gene is associated with coronary atherosclerosis. Circulation. 1995;91:951-954.
Hopkins P, Lifton RP, Hollenberg NK, Jeunemaitre X, Hallouin M-C, Skuppin J, Williams CS, Dluhy RG, Lalouel J-M, Williams RR, Williams GH. Blunted renal vascular response to angiotensin II is associated with a common variant of angiotensinogen and obesity. J Hypertens. 1996;14:199-207.
Williams GH, Dluhy RG, Lifton RP, Moore TJ, Gleason R, Williams R, Hunt SC, Hopkins PN, Hollenberg NK. Non-modulation as an intermediate phenotype in essential hypertension. Hypertension. 1992;20:788-796.