Angiotensinogen in Human Essential Hypertension
The candidacy of angiotensinogen for a role in the genetic basis of hypertension is supported by the observation that plasma angiotensinogen levels track with raised blood pressure through families. In addition, transgenic mice with overexpression of a rat angiotensinogen gene develop hypertension, and knockout mice with a disrupted gene and absent angiotensinogen production develop low blood pressure. There are now two studies in populations of white European origin and one in African Caribbeans providing support for a role of the angiotensinogen gene locus in human essential hypertension.
The angiotensinogen gene is localized to the chromosomal region 1q42-43 and comprises five exons and four introns spanning 13 kb.1 The gene product is a 55- to 60-kD globular glycoprotein that acts as the sole substrate for renin.1 Even though there is a large amount of angiotensinogen in the circulation, it does not constitute an excess substrate for renin cleavage. Indeed, the rate of angiotensin I generation by renin appears to proceed at approximately half the maximal velocity, indicating considerable potential reserve in this reaction.1 Accordingly, an increase in angiotensinogen concentration will lead to increased angiotensin I production even if the plasma renin concentration does not change.
Stimulation of angiotensinogen production occurs over hours or days in response to stimulatory effects of angiotensin II, estrogens, thyroid hormones, and glucocorticoids on regulatory elements in the gene promoter.1 Paradoxically, angiotensin II exhibits positive regulation on angiotensinogen production while exerting a negative feedback influence on renin.
Angiotensinogen in Essential Hypertension
Walker et al2 described a strong positive correlation of angiotensinogen with recumbent diastolic pressure in 541 ambulatory subjects who were not receiving oral contraceptives or any treatment that might affect blood pressure (BP). This same relationship was noted in both white and black Americans and was particularly marked with diastolic pressures greater than 90 mm Hg. The observations were not explained by renin suppression, because a stronger negative correlation of plasma renin activity with BP would be anticipated. By using multiple regression analysis, Walker et al concluded that 15% to 20% of the BP variance could be explained by components of the renin-angiotensin system.2
Further support for an elevation of angiotensinogen with BP emerges from the Ladywell study,3 which accumulated BP measurements in 864 young adults and their parents. This study took a novel approach of identifying four groups from the four corners of a scattergram based on parental and offspring BP scores. These four groups included parents and offspring with higher BPs, parents with low and offspring with high BPs, both parents and offspring with low BP, and parents with high and offspring with low BPs.3 High BPs in the offspring were significantly associated with higher angiotensinogen, cortisol, and 18-hydroxycorticosterone when parental BP was high. Interestingly, the highest angiotensinogen levels were in the groups with both parents and offspring with higher pressures, which suggests that higher levels of substrate track with higher pressures through families. Angiotensin II and atrial natriuretic peptide were similarly, but not significantly, raised with higher BP. The study design means that the angiotensinogen level in offspring correlates with BP and that this relationship depends on parental BP measured 8 years earlier. These data provide support for a role of genetic determination of angiotensinogen in BP regulation.
Angiotensinogen in Experimental Models
In 1992, Kimura et al4 generated hypertensive transgenic mice using the rat angiotensinogen gene. The host mouse strains were particularly selected for their low native angiotensinogen plasma levels and the ability of mouse renin to cleave rat angiotensinogen. In one of the transgenic mouse models, angiotensinogen was overexpressed in an appropriate rat tissue distribution in brain, liver, and vascular tissue. This was associated with raised plasma angiotensinogen and angiotensin II and the development of hypertension. Intriguingly, another transgenic line remained normotensive, despite raised angiotensinogen in the plasma. This was in the context of an aberrant pattern of angiotensinogen expression (compared with that expected for rats) with low levels in brain and liver. These data provide an important demonstration that a transgenic animal model may develop hypertension in association with overexpression of the angiotensinogen gene, although that pattern of tissue expression may be important in determining effects on BP.4
In the context of these experimental observations, it is of particular interest that an angiotensinogen knockout mouse model with undetectable angiotensinogen mRNA in the liver or angiotensinogen peptide in the plasma demonstrated hypotension. Kim et al5 have recently added up to four extra copies of the angiotensinogen gene to transgenic mice models that display incremental increases in BP with each extra copy of this gene. These exciting observations emphasize the role of this substrate in BP regulation.
In contrast, a study of the stroke-prone spontaneously hypertensive rat has not demonstrated linkage of the angiotensinogen gene to hypertension with the use of a PvuII restriction site within exon 2.6 However, problems have arisen previously with attempts to translate very promising genetic studies in animals to humans, which may reflect the relatively low power of many human studies to exclude any gene for hypertension. An alternative explanation might be that animals offer specific refined BP phenotypes arising from homogeneous crosses, whereas studies in humans approach the phenotype of essential hypertension, which is far more heterogeneous in origin.
Linkage of the Angiotensinogen Gene to Hypertension in White Europeans
In the investigation of the genetic basis of hypertension, it is important to consider the variable age of onset (typically late), the incomplete penetrance of susceptibility genes, and the possibility that genetic influences may not obey simple mendelian modes of inheritance.7 These features have encouraged researchers to focus on affected hypertensive sibling pairs by testing whether they share alleles at the angiotensinogen locus more often than would be expected by chance in the general population. This form of linkage analysis centers around computation of a t statistic based on allele sharing that measures excess allele sharing or the difference between the observed sharing of angiotensinogen alleles in affected sibling pairs and the expected sharing of alleles derived from a random control population. This t statistic is sensitive to the numbers analyzed and is not a direct measure of the strength of linkage.7 When small numbers of control subjects are used, the sample size might distort the control allele frequencies, which could inflate or diminish support for linkage. In the analysis of the angiotensinogen locus in white Europeans, the UK and French control allele distribution is very similar, suggesting that population distortions are unlikely to be responsible for the published data.
In 1992, Jeunemaitre et al8 used this method to report the first suggestion that the angiotensinogen locus might play a role in human essential hypertension. By using a highly polymorphic dinucleotide repeat in the 3′ flanking region of the angiotensinogen gene, they were able to demonstrate some support for linkage in 215 sibships from Utah and Paris (t=2.02, P=.02). However, when the 83 sibships from Paris and the 132 sibships from Utah were tested separately, there was only borderline support for linkage among the Paris families (t=1.71, P=<0.05) and none among the Utah families (t=1.22, P=.11). This disparity between the two centers may reflect the fact that the French families were more severely affected hypertensive individuals and had leaner body mass indexes.8 The evidence for linkage was more impressive when sibships were stratified into those that were more severely affected as defined by an arbitrary threshold of diastolic BP greater than 100 mm Hg (110 combined Utah and Paris sibships, t=3.40, P≤.001). Furthermore, there was support for linkage to angiotensinogen among the 171 sibships with onset of hypertension before 45 years of age (t=2.23, P=.02), but partition of the sibships according to sex revealed support for linkage among only male-male pairs.8
In 1994, we reported confirmation of linkage and association of the angiotensinogen locus to hypertension in 63 white European families from the United Kingdom (t=5.0, P=.0000003).9 The striking feature of this study was the consistent support for linkage of the angiotensinogen locus to hypertension in both sexes and in a group of more severely affected families.9 We now present data on an additional 14 white European families that support our previous observations of strong support for linkage of angiotensinogen to hypertension. The demographic data on 77 families are summarized in Table 1⇓. We used the affected pedigree member method of allele sharing to test for linkage of this locus to hypertension. This form of linkage analysis uses weighting scales that recognize that excess sharing of rare alleles among siblings is more remarkable. Interestingly, in these 77 families, there is strong support for linkage even without the application of any weighting scale (Table 2⇓). This confirmation of linkage is particularly encouraging, because it has proved difficult to replicate support for linkage in other complex traits such as schizophrenia.7 Such difficulties have probably arisen from differences in phenotypic classification or may have resulted from the genetic heterogeneity of such traits and the unwitting recruitment of phenocopies or even variable penetrance.7
Together with the studies on French and Utah affected sibships, these data confirm that recruiting families according to the severity of affectation and, when possible, minimizing environmental influences by including lean individuals may be important in the study of the role of angiotensinogen.8 9 This is supported by the observation that families from Utah with a much higher body mass index do not show support for linkage of angiotensinogen to hypertension as a separate population until stratified according to severity and combined with French families.8 This may have implications for other studies and may even suggest that obese hypertensive individuals constitute a distinct genetic group of hypertensive individuals.
Angiotensinogen in African Caribbeans
The high prevalence of hypertension in populations of West African ancestry and observations of higher rates of stroke and renal disease have led to speculation that hypertension may have a different genetic basis among these groups.10 African Caribbean hypertensive individuals have low-renin, sodium-sensitive BP and in some studies respond less well to treatment with angiotensin-converting enzyme inhibitors. It is possible that the apparently low plasma renin activity is a general feature of populations of this ethnicity and not a specific phenomenon associated with a suppressed plasma renin-angiotensin system in those individuals with essential hypertension.10
We have studied the same angiotensinogen marker in 63 African Caribbean families from St Vincent and the Grenadines and have reported some support for linkage and association of this locus to hypertension that implies that there may be similarities in the genetic basis of hypertension between different ethnic groups.11 Since support for linkage arises from 14 of the 63 African Caribbean families who share rare alleles in excess, it is important to remain cautious until these results have been replicated in other populations of West African origin.11
Molecular Variants of Angiotensinogen
The distance over which linkage may be detected between a genetic marker and a susceptibility variant means that these studies may hint at a role for angiotensinogen or indeed an unidentified locus within close proximity. To define a relationship to the angiotensinogen gene, Jeunemaitre et al8 identified 15 molecular variants and used them in case-control studies to test for association with hypertension. Of these polymorphisms, one encoding threonine instead of methionine at position 235 (M235T) and one encoding methionine instead of threonine at position 174 (T174M) in the gene product were associated with hypertension and also with plasma angiotensinogen levels in white Europeans and Americans.8
The variants M235T and T174M are at some distance from both the angiotensin cleavage sites and the promoter region, and it remains uncertain how these polymorphisms might functionally influence the renin-angiotensin system. It appears most likely that these variants are in linkage disequilibrium with another variant that might influence angiotensinogen levels. This is plausible because there are several conflicting studies on association of M235T or T174M with hypertension in populations of white European ancestry, although we have not studied these variants further in the additional families reported here.9 12 13 In addition, studies have reported no association of the angiotensinogen variant M235T with hypertension in African Americans or African Caribbeans from St Vincent.11 14 However, studies in Japanese have demonstrated association of T235 with hypertension, although one of these studies may have some selection bias because of the choice of university staff and students as the control group.15 16
Future Research on Angiotensinogen
Although there is support for linkage of the angiotensinogen locus to hypertension from two studies on populations of white European ancestry and one of African Caribbean origin, the precise variants that influence the relationship of the angiotensinogen locus with hypertension in white European and African Caribbean populations remain to be determined. To resolve the role of angiotensinogen, European collaboration aims to create a large resource of affected sibling pairs that we hope will confirm our previous observations on smaller numbers. A second strategy seeks to define the precise role of new and previously identified variants of the angiotensinogen gene in a large prospective epidemiological study of cardiovascular risk. This will afford researchers an opportunity to study how BP tracks with angiotensinogen variants over time and investigate relationships of this to plasma levels of angiotensinogen.
This work was supported by The Joint Research Board of St Bartholomew's Hospital, The Mason Medical Foundation, The Fellowship of Postgraduate Medicine, and The Wellcome Trust. We thank the people of St Vincent and the Grenadines and the United Kingdom who assisted with these studies.
Reprint requests to Mark Caulfield, Department of Clinical Pharmacology, St Bartholomew's Hospital, London, EC1A 7BE, UK.
- Received June 29, 1996.
- Revision received July 29, 1996.
- Accepted July 29, 1996.
Menard J, Clauser E, Bouhnik J, Corvol P. Angiotensinogen: biochemistry. In: Robertson JIS, Nicholls MG, eds. The Renin-Angiotensin System. London, UK: Mosby; 1993:8.1-8.10.
Walker WG, Whelton PK, Saito H, Russell RP, Hermann J. Relationship 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 GC, Harrap SB, Foy CJ, Holton DW, Edwards HV, Davidson HR, Connell 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.
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.
Lander ES, Schork NJ. Genetic dissection of complex traits. Science. 1994;265:2037-2048.
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;95:687-692.
Rotimi C, Morrison L, Cooper R, Oyejide C, Effiong E, Ladipo M, Osotemihen B, Ward R. Angiotensinogen gene in human hypertension: lack of an association of the 235T among African Americans. Hypertension. 1994;24:591-594.
Hata A, Namikawa C, Sasaki M, Sato K, Nakamura T, Tamura K, Lalouel JM. Angiotensinogen as a risk factor for essential hypertension in Japan. J Clin Invest. 1994;93:1285-1287.
Iwai N, Shimoike H, Ohmichi N, Kinoshita M. Angiotensinogen gene and blood pressure in the Japanese population. Hypertension. 1995;25:688-693.