(Hypertension. 1999;33:1052-1056.)
© 1999 American Heart Association, Inc.
Scientific Contributions |
Association of an Apolipoprotein B Gene Marker With Essential Hypertension
From the Departments of Pathology (P.M.F.) and Internal Medicine (E.N.O.), Faculty of Medicine & Health Sciences, United Arab Emirates University, Al Ain; and the Department of Internal Medicine, Tawam Hospital (G.G.L.), Al Ain, United Arab Emirates.
Correspondence to Dr Philippe M. Frossard, Department of Pathology, Faculty of Medicine & Health Sciences, PO Box 17666, Al Ain, United Arab Emirates. E-mail frossard{at}emirates.net.ae
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract—We designed an association (retrospective, case control) study aimed at evaluating associations between genetic variations of the human apolipoprotein B (apoB) gene and clinical diagnosis of essential hypertension. Our approach was to compare the distribution of the alleles of a highly polymorphic variable number of tandem repeats localized 3' to the human apoB gene, the apoB 3' hypervariable region (HVR), in a group of normotensive and a group of hypertensive individuals. We collected DNA samples from 437 unrelated nationals (215 normotensives and 222 hypertensives) from the United Arab Emirates (UAEs), and we determined their apoB 3' HVR allele and genotype status with a polymerase chain reaction–based assay. In the UAE population, we found 18 alleles underlying a total of 51 genotypes. The distribution of these alleles was significantly different between normotensive and hypertensive UAE nationals. The main peak of the distributions occurred at 35 repeats among hypertensives (with a relative frequency of 25.7% versus 19.6% in normotensives) and at 37 repeats among normotensives (28.8% versus 20.3% in hypertensives). Alleles with 21, 23, 25, 49, and 55 repeats were found in hypertensives only (with a combined relative frequency of 7.6%). We conclude that variations of the apoB gene, or of a nearby gene, that may be in linkage disequilibrium with these alleles play a role in the development of essential hypertension in the UAEs.
Key Words: apolipoproteins B • case-control studies • genetics • hypertension, essential • repetitive sequences, nucleic acid • United Arab Emirates
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The aim of molecular geneticists with respect to the field of hypertension is to identify the part played by genetic factors in the onset of the disorder. The first step is to uncover quantitative trait loci (QTLs) whose products are implicated in abnormal determination of blood pressure.1 2 3 Among the many candidate genes suspected to be involved in human essential hypertension, the apolipoprotein B (apoB) gene is of particular interest because of its demonstrated association with cardiovascular diseases (CVDs).4 A highly polymorphic genetic marker has been described 73 bp 3' to the second polyadenylation signal of the human apoB gene, which is located on chromosome 2 and is commonly referred to as apoB 3' HVR (hypervariable region).5 6 Several alleles of this polymorphic system have been found to be associated with coronary heart disease (CHD) and myocardial infarction as well as with various hyperlipidemia in different populations.4 7 8 9
A number of clinical entities, such as hypercholesterolemia, hypertension, insulin-resistance syndromes, and obesity, participate in an intricate interaction in the development of atherosclerosis and CVDs. Common genetic susceptibilities probably underlie the phenotypic expression of these classic risk factors. Therefore, candidate genes for a specific clinical entity involved in CVDs, such as atherosclerotic heart disease, are also good candidates for unraveling the molecular architecture of another implicating clinical phenotype, such as hypertension. Yet, to our knowledge, despite the key role played by apoB in the molecular etiology of CVDs, only 1 molecular genetic study has attempted to observe putative correlations of apoB genetic markers and essential hypertension, and it failed to uncover an association between apoB 3' HVR and primary hypertension in a Japanese population.10
Genetic drift and admixture are usually confounding factors in the unraveling of the genetic architecture of complex (multifactorial) traits. On the basis of this reason and also to minimize the problems that underlie investigations based on the concept of linkage disequilibrium, it is of utmost importance to investigate genetically homogeneous ethnic groups.11 12 13 We report here the results of an association (retrospective, case control) study aimed at evaluating the relationship of the apo B 3' HVR with essential hypertension. This study was performed on nationals from the Abu Dhabi Emirate, a genetically homogeneous ethnic population without a history of smoking or alcohol consumption.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
The United Arab Emirates (UAEs) is a federation of 7 Emirates, the Abu Dhabi Emirate is the largest, with an indigenous population that comprises UAE nationals, who are Gulf Arabs of Bedouin descent. Until recently, Gulf Bedouin tribes have been characterized by population migrations that have been restricted to this area: intratribal marriages were usual and consanguinity levels were high. The incidence of essential hypertension in the UAEs is similar to that in other parts of the world.14 In this pilot, retrospective case-control study, we investigated a sample population of 437 UAE nationals (Emirati) from the Abu Dhabi Emirate for putative associations between the apoB 3' HVR and essential hypertension in a genetically homogeneous ethnic group. This project was approved by the Research Ethics Committee of the Faculty of Medicine and Health Sciences (UAE University, Al Ain, UAE), and all subjects gave their informed consent.
The sample population of 437 unrelated subjects (225 men, 212 women) had a mean age±SD of 52.5±13.1 years and was composed of the 2 following groups: 222 patients with primary hypertension (115 men, 107 women; age, 52.7±12.1 years; total serum cholesterol, 7.0±2.7 mmol/L; body mass index (BMI), 30.1±6.4), and 215 "controls" used as a comparison group (110 men, 105 women; age, 51.5±13.3 years; total serum cholesterol, 5.2±0.5 mmol/L; BMI, 29.2±6.5). Total serum cholesterol values were taken as off lipid-lowering medication values from subjects medical charts. Serum apoB and low-density lipoprotein level determinations are not part of routine investigations, and these values were thus unavailable from patient charts. No member of the sample populations admitted to alcohol intake, and there was no history of smoking in any of the subjects.
Patients were classified as having essential hypertension if they had systolic blood pressures >140 mm Hg and diastolic blood pressures >90 mm Hg on at least 3 separate occasions; had no clinical signs, symptoms, and laboratory findings suggestive of secondary hypertension; and had positive family history of hypertension (as assessed by direct questioning of occurrence of hypertension in any close relative). These subjects were all known hypertensive patients who were recruited for this study during ambulatory visits to the hypertension unit for routine checkups. All hypertensive subjects were on antihypertensive medications, and therefore blood pressure values could not be taken into account for this study.
"Control" (comparison) subjects were healthy individuals who had no personal history of hypertension (documented by individual medical records as resting systolic blood pressures <140 mm Hg and diastolic blood pressures <90 mm Hg on at least 3 separate occasions) and no family history of hypertension. Control subjects were on neither antihypertensives or any other medication that could affect blood pressure values. Control individuals were taken from a DNA database that we are building from a random population of unrelated UAE nationals (it is currently composed of 3100 DNA samples); they were recruited as normotensive individuals whose age, gender, and BMI matched those of the hypertensive group.
Care was taken to ensure that no subject of this study (whether in hypertensive or normotensive groups) was affected by confounding clinical phenotypes, including non–insulin-dependent diabetes mellitus (that is otherwise quite prevalent in this population) and ischemic heart disease. Moreover, all hypertensive patients had absence of left ventricular hypertrophy as assessed by echocardiography.
DNA Analysis
DNA was extracted from peripheral blood
leukocytes isolated from 5-mL blood samples drawn for routine clinical
investigations according to standard protocols, and genotypes
of the apoB 3' HVR were determined by polymerase chain reaction using
the method of Boerwinkle et al.6 Because agarose gel
electrophoresis may lead to equivocal assignment of HVR
alleles,15 polymerase chain reaction products were
visualized after electrophoretic migration in 4%
polyacrylamide gels and staining with ethidium bromide.
On average, the human apoB 3' HVR is constituted by tandem repeats of a 30-bp core DNA sequence.6 The usual nomenclature to describe this polymorphism is to take into account the average number of core, consensus sequences, which are referred to as hypervariable elements (HVE). Thus, a 35-time repeat of the core sequence, for example, is called HVE35, as indicated in the nomenclature of Boerwinkle et al.6
Data Analyses
Allele frequencies were determined by the method of gene
counting.16 Statistical determinations were performed with
the help of an SPSS version 6.1 for Windows software package
(Gorinchem). Differences in the distributions of apoB 3' HVR
alleles on the basis of clinical phenotypes (normotensive
versus hypertensive) were assessed with 
2
procedures on corresponding tables of association. Odds ratio (OR)
values (95% CI) were determined directly from 2x2 table associations.
For all analyses, statistical significance was considered at
P<0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study included 437 Emirati subjects (225 men, 212 women) with complete phenotypic and genotypic data. Normotensive "controls" represented a "comparative" rather than a "control" group; they were indeed free of disease up to the time of investigation. They also had no family history of hypertension, which was ascertained by direct questioning and by readouts of individual medical records. They were included in the protocol because their BMIs, age, and gender matched those of the hypertensive patients. Their serum cholesterol levels were significantly lower than those of the hypertensive patients (5.2±0.5 versus 7.0±2.7 mmol/L, P<0.05). The control subjects, who had neither a personal or family history of essential hypertension, represented a valid comparison group for this association study. BMI values indicate that all subjects were overweight (26<BMI<30) and that hypertensives were even slightly obese (BMI>30). These BMI values are representative of the UAE population, which is an affluent and rapidly expanding society: being overweight is a general characteristic of its members (P.M.F., unpublished data, 1998).
Table 1 displays the distribution of the different alleles observed between the 2 groups of 215 normotensive and 222 hypertensive UAE nationals. In the overall population of 437 subjects, we observed the presence of 18 alleles, which ranged from HVE21 to HVE55, and of 51 genotypes (data not shown). In normotensives, the distribution of 13 alleles followed a quadrimodal distribution (Table 1) with peaks at HVE31 (respective allele frequency, 15.3%), HVE37 (28.0%), HVE47 (7.7%), and HVE51 (2.8%). In hypertensives, the distribution of 18 alleles gave a trimodal distribution (Table 1) with peaks at HVE31 (16.9%), HVE35 (25.7%), and HVE47 (6.1%).
|
To compare allele frequency distributions in different groups and
different populations, a 4-allele model has been
proposed,17 in which apoB 3' HVR alleles are
registered into 4 classes: HVE<35, HVE35, HVE37, and HVE>37 (Table 2). Allele frequency distribution
difference between the groups of normotensive and hypertensive subjects
was assessed in the 4x2 table of association (Table 2) by a
2 test with 3 df. The test result was
2=10.11, P=0.017. HVE35 were
therefore associated with clinical diagnosis of hypertension (25.7%
among hypertensives versus 19.6% in normotensives), whereas HVE37 were
associated with normotensive status (28.8% in normotensives versus
20.3% in hypertensives). The OR of the association of HVE35 versus
HVE37 with essential hypertension is 1.85 (Cornfield 95% CI for OR are
1.23 to 2.80). Furthermore, HVE36 (0.7%) occurred among only
normotensive subjects.
|
To document differences in the distribution of less common alleles,
we designed an a posteriori model in which alleles smaller than
HVE35 and larger than HVE37 were pooled to illustrate complete absence
of some alleles in the group of normotensives. This model is
displayed in Table 3, and
2 analysis results on the 8x2 table
of association were 2=48.29, 7 df,
P<1x10-8. Tables 1 and 3 show that HVE21, HVE23, HVE25, HVE49, and HVE55 occurred in
the group of hypertensive subjects only (at relative frequencies of
1.4%, 1.4%, 0.7%, 3.4%, and 0.7% respectively), whereas HVE51 were
more frequent among normotensive individuals (2.8% versus 0.7%).
Statistical differences existed between normotensives and hypertensives
in the distribution of all alleles except for HVE27 to HVE33 and
HVE39 to HVE47 (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
It is now well established that association studies at candidate gene loci represent a powerful approach to detect small genetic effects and thus to identify QTLs with reduced or low penetrance; however, these methods are influenced by the effects of population stratification, selection bias, confounding by other variables, and clinical definitions used to characterize patient groups.11 12 13 One way to minimize the weight of the first 2 issues is to probe putative associations between candidate gene variations and a complex, heterogeneous disorder such as primary hypertension in ethnic groups that may be genetically more homogeneous. In such cases, disease-causing mutations are present on 1 or a few ancestral haplotypes and linkage disequilibrium occurs over greater genetic distances than in genetically heterogeneous (or older) populations. The population of UAE subjects that we recruited for this study offered another advantage in that there was no history of smoking or alcohol intake, 2 of the usual confounding environmental factors.
In this study, we demonstrate for the first time a marked, statistically significant association of apoB 3' HVR alleles and direct clinical diagnosis of essential hypertension. Several studies have demonstrated correlations between genetic variations of the apoB gene and atherosclerotic CHD and dyslipoproteinemia.7 18 19 Such polymorphisms include apoB 3' HVR. From the time that Hegele et al20 reported an association of the larger apoB 3' HVR alleles with myocardial infarction, similar associations with myocardial infarction and with CHD were confirmed in other populations.8 21 22 Another study established an association with the smaller apoB 3' HVR.23 Other research groups have reported negative findings.24 25
Despite controversial data in the field of hypertension, there is general agreement that apoB variants that may be in linkage disequilibrium with some apoB 3' HVR alleles are directly implicated in dyslipoproteinemia and allied CHD. This working hypothesis has recently been clearly demonstrated by Tybjærg-Hansen et al,26 who have shown that mutation Arg3500Gln in the apoB gene causes severe hypercholesterolemia and increases the risk of ischemic heart disease and that this mutation could form the basis for a screening test in the white population in Denmark. More detailed studies aimed at relating apoB gene locus polymorphisms and CVDs should therefore contribute to evidence associations with clinical and intermediate phenotypes that underlie an individual's genetic susceptibility to CVDs.
Clear clinical, epidemiological, and genetic evidence exists that indicates dyslipidemia accompanies essential hypertension and that biological interrelations between blood pressure and blood lipids may influence the mechanisms that associate hypertension and CHD.27 28 29 30 Thus, studies that focus on unraveling the genetic architecture (based on apoB variants) of coronary artery disease, non–insulin-dependent diabetes mellitus, and hypertension are gaining acceptance.4 7 26 However, in a preliminary investigation on a restricted Japanese population, Higashimori et al10 reported no association between apoB 3' HVR and essential hypertension.
We have shown in a population of UAE nationals that the common allele HVE35 is significantly associated with clinical diagnosis of hypertension: it represents the most frequent allele among hypertensives, in whom its relative frequency is 25.7% (compared with 19.6% in normotensives). HVE37, on the other hand, is the most common allele in normotensives (28.8% versus 20.3% in hypertensives). In other populations that have been studied worldwide, the main peak of apoB 3' HVR allelic distribution occurs at either HVE35 or at HVE37,17 with a marked frequency difference between the two. In Emirati, HVE35 and HVE37 occur with similar frequencies in a random population of subjects (Frossard, unpublished data, 1998), and a difference becomes apparent when investigating either hypertensive or normotensive subjects. Furthermore, less common individual alleles HVE21, 23, 25, 49, and 55 are present among hypertensives only; taken together, these alleles represent 7.6% of all alleles found in hypertensives (Table 1). It is interesting to note that "minor" alleles such as HVE25 and HVE60 were also found only in coronary artery disease and stroke patients from Taiwan.9 Therefore, apoB variants, or variants of any nearby gene, that may be in linkage disequilibrium with HVE21, HVE23, HVE25, HVE35, HVE49, and HVE55 contribute significantly to the development of hypertension in the Emirati population.
Thus, the apoB gene, or any gene that may be in linkage disequilibrium with it, constitutes a good candidate for participation in the determination of an individual's genetic susceptibility to essential hypertension in the indigenous population of the UAE. However, association does not imply causation, and such results should be interpreted with care.31 Nonetheless, the conclusions reached in this report warrant testing in other populations.
Received September 1, 1998; first decision October 7, 1998; accepted December 11, 1998.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Ward R. Familial aggregation and genetic epidemiology of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press; 1990:81–100.
-
Lifton RP. Genetic determinants of human hypertension.
Proc Natl Acad Sci U S A. 1995;92:8545–8551.
[Abstract/Free Full Text] - Soubrier F, Lathrop GM. The genetic basis of hypertension. Curr Opin Nephrol Hypertens. 1995;4:177–181.[Medline] [Order article via Infotrieve]
- Humphries SE, Dunning A, Xu CF, Peacock R, Talmud P, Hamsten A. DNA polymorphism studies: approaches to elucidating multifactorial ischemic heart disease: the apo B gene as an example. Ann Med. 1992;24:349–356.[Medline] [Order article via Infotrieve]
-
Knott TJ, Wallis SC, Pease RJ, Powell LM, Scott J. A
hypervariable region 3' to the human apolipoprotein B gene.
Nucleic Acids Res. 1986;14:9215–9216.
[Free Full Text] -
Boerwinkle E, Xiong W, Fourest E, Chan L. Rapid typing
of tandemly repeated hypervariable loci by the polymerase chain
reaction: application to the apolipoprotein B 3' hypervariable
region. Proc Natl Acad Sci U S A. 1989;86:212–216.
[Abstract/Free Full Text] - Vedie B, Myara I, Jeunemaitre X, Moatti N. Genetic variations in the apolipoprotein B gene (in French). Ann Génét. 1995;38:187–201.[Medline] [Order article via Infotrieve]
- Ye P, Chen B, Wang S. The association of polymorphisms at a VNTR locus 3' to the apolipoprotein B gene with coronary heart disease in Chinese population. Chinese Med Sci J. 1995;10:63–69.
-
Wu JH, Chern MS, Lo SK, Wen MS, Kao JT. Apolipoprotein
B 3' hypervariable repeat genotype: association with plasma
lipid concentration, coronary artery disease, and other
restriction fragment polymorphisms. Clin Chem. 1996;42:927–932.
[Abstract/Free Full Text] - Higashimori K, Higaki J, Miki T, Kamitani A., Mikami H, Kumahara Y, Ogihara T. Analysis of the apolipoprotein B3' hypervariable region in patients with essential hypertension. Clin Exp Pharmacol Physiol. 1992;19(suppl 20):21–23.
- Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science. 1996;273:1516–1517.[Medline] [Order article via Infotrieve]
-
Lander ES. The new genomics: global views of biology.
Science. 1996;274:536–539.
[Free Full Text] - Schork N. Genetically complex cardiovascular traits: origins, problems, and potential solutions. Hypertension. 1997;29(pt 2):145–149.
- Frossard PM, Obineche EN, Elshahat YI, Lestringant GG. Deletion polymorphism in the angiotensin-converting enzyme gene is not associated with hypertension in a Gulf Arab population. Clin Genet. 1997;50:211–213.
- Latorra D, Stern CM, Schanfield MS. Characterization of human AFLP systems apolipoprotein B, phenylalanine hydroxylase, and D1S80. PCR Methods Appl. 1994;3:351–358.[Medline] [Order article via Infotrieve]
- Nei M. Molecular evolutionary genetics. Columbia University Press; New York, NY: 1987.
- Alavantic D, Glisic S, Erceg S, Stupar M. Genetic variation at the apoB 3' hypervariable region in a Serbian population. Eur J Hum Genet. 1997;5:333–335.[Medline] [Order article via Infotrieve]
- Hegele RA, Breslow JL. Apolipoprotein genetic variation in the assessment of atherosclerosis susceptibility. Genet Epidemiol. 1987;4:163–184.[Medline] [Order article via Infotrieve]
-
Paulweber B, Friedl W, Krempler F, Humphries SE,
Sandhofer F. Association of DNA polymorphism at the apolipoprotein
B gene locus with coronary heart disease and serum very low
density lipoprotein levels.
Arteriosclerosis. 1990;10:17–24.
[Abstract/Free Full Text] - Hegele RA, Huang L, Herbert PN, Blum CB, Buring JE, Hennekens CH, Breslow JL. Apolipoprotein B gene DNA polymorphisms associated with myocardial infarction. New Engl J Med. 1986;315:1509–1515.[Abstract]
- Friedl W, Ludwig EH, Paulweber B, Sandhofer F, McCarthy BJ. Hypervariability in a minisatellite 3' of the apolipoprotein B gene in patients with coronary heart disease compared with normal controls. J Lipid Res. 1990;31:659–665.[Abstract]
- Tybjærg-Hansen A, Nordestgaard BG, Gerdes LU, Humphries SE. Variation of apolipoprotein B gene is associated with myocardial infarction and lipoprotein levels in Danes. Atherosclerosis. 1991;89:69–81.[Medline] [Order article via Infotrieve]
- Deeb S, Failor A, Brown BG, Brunzell JD, Albers JJ, Motulsky AG. Molecular genetics of apolipoprotein and coronary heart disease. Cold Spring Harb Symp Quant Biol. 1986;51(pt 1):403–409.
- Tikkanen MJ, Heliö T. Genetic variants of apolipoprotein B: relation to serum lipid levels and coronary artery disease among the Finns. Ann Med. 1992;24:357–361.[Medline] [Order article via Infotrieve]
- Turner PR, Talmud PJ, Visvikis S, Ehnholm C, Tiret L. DNA polymorphisms of the apoprotein B gene are associated with altered plasma lipoprotein concentrations but not with perceived risk of cardiovascular disease: European Atherosclerosis Research Study. Atherosclerosis. 1995;116:221–234.[Medline] [Order article via Infotrieve]
-
Tybjærg-Hansen A, Steffensen R, Meinertz H, Schnohr P,
Nordestgaard BG. Association of mutations in the apolipoprotein B gene
with hypercholesterolemia and the risk of
ischemic heart disease. N Engl J Med. 1998;338:1577–1584.
[Abstract/Free Full Text] -
Bonaa KH, Thelle DS. Association between blood pressure
and serum lipids in a population: the Tromso Study.
Circulation. 1991;83:1305–1314.
[Abstract/Free Full Text] - Kistler T, Weisser B. Correlation between disorders of lipid metabolism and hypertension in 10,892 participants in the Heureka Study [in German]. Schweiz Rundsch Med Prax. 1993;82:1222–1233.[Medline] [Order article via Infotrieve]
- Williams RR, Hunt SC, Hopkins PN, Wu LL, Hasstedt SJ, Berry TD, Barlow GK, Schumacher MC, Ludwig EH. Genetic basis of familial dyslipidemia and hypertension: 15-year results from Utah. Am J Hypertens. 1993;6(pt 2):319S–327S.
-
Caulfield M, Bouloux PM, Munroe P. Progress in
determining the genes for hypertension, insulin resistance, and
dyslipidemia. Ann N Y Acad Sci. 1997;827:110–117.
[Abstract/Free Full Text] -
Re RN, Frohlich ED. Controversies in the genetic
analysis of hypertensive diseases. Hypertension. 1996;28:880. Editorial.
[Free Full Text]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||