Author + information
- Received August 22, 1997
- Revision received April 8, 1998
- Accepted April 23, 1998
- Published online August 1, 1998.
- Kazuomi Kario, MD, PhD∗,*,
- Takefumi Matsuo, MD, PhD‡,
- Hiroko Kobayashi, PhD§,
- Nobuyuki Kanai, MD, PhD†,
- Satoshi Hoshide, MD∗,
- Takeshi Mitsuhashi, MD, PhD∗,
- Uichi Ikeda, MD, PhD∗,
- Shinichi Nishiuma, MD∥,
- Masafumi Matsuo, MD, PhD∥ and
- Kazuyuki Shimada, MD, PhD∗
- ↵*Address for correspondence: Dr. Kazuomi Kario, MD, PhD, Department of Cardiology, Jichi Medical School, 3311-1, Yakushiji, Minamikawachi, Kawachi, Tochigi, 329-0498, Japan
Objectives. The purpose of this study was to investigate the angiotensin-converting enzyme (ACE) insertion/deletion (I/D) genotype and endothelial cell dysfunction or hypercoagulable state in elderly hypertensive patients.
Background. Angiotensin-converting enzyme (ACE) insertion/deletion (I/D) polymorphism was recently reported to be associated with various cardiovascular diseases. However, the precise mechanism of this association remains unknown, and some confounding factors might also affect the association. Endothelial cell dysfunction and coagulation activation play important roles in both the atherosclerotic process and the onset of cardiovascular events.
Methods. We identified the ACE I/D genotype and measured the plasma levels of markers of endothelial cell damage (von Willebrand factor [vWF] and thrombomodulin) and of coagulation activation (prothrombin fragment F1 + 2 [F1 + 2]) in 318 asymptomatic elderly patients with hypertension, aged 59–93 years.
Results. The vWF level was significantly higher in those with the DD genotype (n = 54) than in those with the II genotype (n = 131, p < 0.0001) or with the ID genotype (n = 133, p < 0.0001). The TM levels were also higher in patients with the ID genotype (p < 0.005) and the DD genotype (p < 0.01) than in those with the II genotype. There were no differences in F1 + 2 level among the groups. Positive correlations of systolic blood pressure with levels of both vWF and thrombomodulin were found predominantly in patients with the II genotype (both p < 0.001), but no correlation was noted in those with the DD genotype.
Conclusions. Considering the increased plasma levels of both endothelial cell-derived markers in the hypertensive patients with ACE DD genotype, we speculate that the ACE D allele is a risk factor for the development of hypertensive cardiovascular disease associated with endothelial cell damage.
Angiotensin-converting enzyme (ACE) plays an important role in vascular homeostasis. ACE acts predominantly as an ectoenzyme of vascular endothelial cells, and is a component of the renin–angiotensin and kallikrein–kinin systems. Recently, ACE insertion/deletion (I/D) polymorphism was reported to be associated with various arterial atherothrombotic diseases in whites and in Japanese, including myocardial infarction, coronary artery spasm, coronary atherosclerosis, hypertrophic and ischemic cardiomyopathy and hypertensive ischemic renal and cerebrovascular disease (1–8). However, the precise mechanism of this association remains unknown, and some confounding factors might also affect the association. Recently, it was shown that the ACE DD genotype is associated not with the progression of coronary or carotid stenosis, but with cardiovascular events (myocardial infarction or lacunar stroke) (7,9,10), indicating that vascular spasm or thrombosis may be an important factor.
Endothelial cell dysfunction and coagulation activation play important roles in both the atherosclerotic process and the onset of cardiovascular events. Some reports indicate that this ACE polymorphism is associated with neither insulin resistance nor abnormality of fibrinolysis (11). A recognized standard for endothelial cell dysfunction is not available, and the relationship between the ACE I/D genotype and endothelial cell dysfunction has not been thoroughly investigated. The plasma levels of von Willebrand factor (vWF) and thrombomodulin have come to be widely used as indicators of endothelial cell dysfunction and damage in vivo (12–14), because they were reported to increase in parallel with the degree of endothelial cell damage both in vitro and in vivo (15–17). Plasma prothrombin fragment 1+2 (F1+2) is a sensitive indicator of coagulation activation, which may reflect hypercoagulable states (18,19). The plasma levels of these molecular markers of endothelial cell damage and coagulation activation are increased in patients with overt and silent cardiovascular diseases (14).
We examined the ACE I/D polymorphism and measured the plasma levels of vWF, thrombomodulin, and F1+2 in elderly hypertensive patients without overt cardiovascular disease to study the relationship of ACE I/D polymorphism to endothelial cell damage and coagulation activation.
We studied 318 of 407 consecutive elderly hypertensive outpatients, aged 59–93 years. Hypertension was diagnosed when the systolic blood pressure (BP) was ≥140 mm Hg and/or the diastolic BP was ≥90 mm Hg (20), or when the patients were receiving antihypertensive therapy. BP was measured in the sitting position using an automated sphygmomanometer (BP103N-II, Nippon Colin, Co., Ltd., Komaki, Japan) after the patient had rested for at least 5 min. We obtained a detailed history regarding the duration of hypertension and cardiovascular complications, such as stroke or coronary artery disease (myocardial infarction and angina pectoris) from each participant, and we also conducted a physical examination and laboratory studies, including blood and urine tests, chest X-ray, and resting electrocardiogram (ECG). We excluded 89 patients because of renal failure, hepatic damage (serum level of creatinine >1.5 mg/dl, blood urea nitrogen >30 mg/dl, or aspartate aminotransferase or alanine aminotransferase >40 IU/liter), suspected diabetes mellitus (fasting glucose >140 mg/dl and/or hemoglobin A1c> 6.4%) or present or previous coronary artery disease, stroke, congestive heart failure, atrial fibrillation or malignancy. The remaining 318 hypertensive subjects met the World Health Organization criteria for stage I or II hypertension, and none of them had factors in their history or routine laboratory test results that suggested secondary hypertension. No antihypertensive therapy was administered in 126 (40%) of the 318 hypertensive patients for at least 1 month before examination. All the remaining treated patients were asked to discontinue their antihypertensive therapy 7 to 14 days before examination, but 70 (22%) of them were not able to do so. Eighty-five subjects (27%) were or had been taking ACE inhibitors. We also studied a control group of 263 healthy normotensive subjects (42% male) who participated in an annual examination whose mean (SD) age was 71 (8.1) years.
Dyslipidemia was diagnosed when the fasting serum cholesterol, triglycerides, and high density lipoprotein (HDL) cholesterol levels were >220 mg/dl, >150 mg/dl and/or <35 mg/dl, respectively. Smokers were defined as current smokers. The body mass index was calculated as weight (kg)/height (m)2. Left ventricular hypertrophy diagnosed by ECG (ECG-LVH) was defined as abnormally high voltages of QRS complexes (R in V5 plus S in V1 greater than 3.5 mV) associated either with flat T waves (less than 10% of R) or with ST segment depression and biphasic T waves (21).
This study was approved by our institutional review committee, and informed consent was obtained from each subject studied.
Polymerase chain reaction for detection of ACE I/D polymorphism
Genomic deoxyribonucleic acid (DNA) was extracted from citrated whole blood using salt/chloroform by a modification of a previously described method (22). Enzymatic amplification of DNA was performed by polymerase chain reaction (PCR) using 1 μg of DNA extract and thermostable Taq polymerase (Takara Biochemical, Kyoto, Japan) by previously described methods (5). All samples were studied independently by one of the authors (N.K.) at the Department of Pathology at Jichi Medical School (Tochigi, Japan). To avoid mistyping, in the samples identified as showing DD genotype by the above method, we conducted an insertion-specific second amplification using the insertion-specific primer of the oligonucleotide sequences of 5′ TGG GAC CAC AGC GCC CGC CAC TAC 3′ and 5′ TCG CCA GCC CTC CCA TGC CCA TAA 3′ (23)and confirmed that there was no misclassification. This confirmation was independently conducted by one of the authors (S.N.) at the Division of Genetics, International Center for Medical Research, Kobe University School of Medicine (Kobe, Japan).
The plasma levels of vWF and thrombomodulin were determined with enzyme-linked immunosorbent assay (ELISA) kits (Diagnostica Stago, Asnieres, France, and Teijin Co. Ltd., Tokyo, Japan, respectively). Fibrinogen levels were measured with a Data-Fi one-stage clotting assay kit (Dade, FL). F1+2 levels were assayed with an ELISA kit (Behringwerke AG). For vWF assay, the value of the pooled plasma, CTS Standard Plasma (purchased from Behringwerke AG, Marburg, Germany), was taken as 100%.
The levels of total cholesterol, triglycerides, glucose, uric acid and creatinine were measured by routine chemical methods. HDL cholesterol was determined by an enzymatic procedure after precipitation with phosphotungstic acid. Lipoprotein(a) was measured with an ELISA (Biopool, Umea, Sweden).
As measured in our laboratory, the coefficient of variation was 2.5% for the fibrinogen assay, 3.8% for the vWF assay, 4.7% for the thrombomodulin assay and 4.2% for the F1+2 assay.
Data are expressed as the mean (SD). Allele frequencies were calculated by gene counting. After one-way analysis of variance (ANOVA), the Scheffé F test was used for the comparison between mean values of two groups for different ACE genotypes. We also used the Kruskal–Wallis test and Mann–Whitney tests as nonparametric tests for the comparison between median values in two groups for different ACE genotypes. Student ttest was used to test differences between the mean values of the two groups of untreated hypertensive subjects and those under antihypertensive treatment. The chi-square test was used to compare the prevalence of risk factors. The differences with a p value <0.05 were considered significant.
Angiotensin-converting enzyme genotype and clinical and metabolic characteristics of the hypertensive patients
Of the 318 hypertensive subjects studied, 131 (41%) showed ACE genotype II, 133 (42%) genotype ID and 54 (17%) genotype DD, and the calculated gene frequencies were 0.62 for I allele and 0.38 for D allele. The clinical and biochemical metabolic characteristics of the subjects with each ACE genotype are shown in Table 1. There were no significant differences among the three groups in the demographic data for age, gender, body mass index, duration of hypertension, smoking status, BP level or lipid profile. There were also no differences among the three groups in the proportion of patients under antihypertensive therapy or of those who were or had been taking ACE inhibitors.
Angiotensin-converting enzyme genotypes and levels of markers of endothelial cell damage and coagulation activation in hypertensive subjects
The levels of the markers of endothelial cell damage (vWF and thrombomodulin) and coagulation activation (F1+2) are shown in Table 2. The vWF level was significantly higher in the patients with DD genotype than in those with II genotype (p < 0.0001) or ID genotype (p < 0.0001). The thrombomodulin levels were significantly higher in those with ID genotype (p < 0.005) and DD genotype (p < 0.01) than in those with II genotype. When a high level of each parameter was defined as >1 SD above the mean, the proportion of the subjects with high vWF level was significantly greater among those with DD genotype than in those with II genotype (p < 0.005), and high thrombomodulin levels were more common in those with the ID genotype (p < 0.05) or the DD genotype (p < 0.001) than in those with the II genotype. There were no significant differences among the three genotype groups in fibrinogen or F1+2 levels.
Severity and duration of hypertension and markers of endothelial cell damage in different ACE genotypes
There were significant positive correlations between BP and markers of endothelial cell damage (vWF and thrombomodulin) in the 318 hypertensive patients, as shown in Table 3. For the three groups with different ACE genotypes, however, positive correlations for these associations were found predominantly in those with the II genotype (Fig. 1). In those with the ID genotype, there were significant correlations of systolic BP with levels of vWF and thrombomodulin (p < 0.01), whereas there were no correlations of BP with any marker of endothelial cell damage in those with the DD genotype (Fig. 1).
There was no significant relationship between the duration of hypertension and the levels of the markers of endothelial cell damage (vWF and thrombomodulin) for either the entire group of 318 patients or for each of the three groups with different ACE genotypes (Table 3).
There was no significant relationship between the BP level or the duration of hypertension and any other metabolic or hemostatic parameter, including fibrinogen and F1+2, for either the entire group or for each of the groups with different ACE genotypes (data not shown).
Antihypertensive therapy and markers of endothelial cell damage in hypertensive subjects
The mean (SD) levels of endothelial cell-derived markers in the untreated hypertensive subjects (n = 248) were not significantly different from those in the subjects under antihypertensive therapy (n = 70) (vWF: 157 [42%] vs. 147 [44%]; thrombomodulin: 28 [12%] ng/ml vs. 26 [12%] ng/ml). There were also no significant differences between the levels of these markers in those taking ACE inhibitor and in those taking other antihypertensive agents (data not shown).
Antihypertensive therapy did not seem to affect any metabolic or hemostatic parameter, including fibrinogen and F1+2, either for the entire group or for each of the groups with different ACE genotypes (data not shown).
Angiotensin-converting enzyme genotypes and markers of endothelial cell damage in normotensive subjects
Of the 263 elderly normotensive subjects studied, 100 (38%) showed ACE genotype II, 133 (51%) genotype ID and 30 (11%) genotype DD, and the calculated gene frequencies were 0.63 for I allele and 0.37 for D allele. Therefore, the ACE genotype distribution and gene frequency in the 263 normotensive subjects were not significantly different from those in the 203 hypertensive subjects. In the normotensive subjects, no significant differences among the groups with different ACE genotypes were found in the levels of either marker of endothelial cell damage (Table 4).
In this study of asymptomatic elderly hypertensive patients, we clarified that the plasma levels of endothelial cell-derived factors (vWF and thrombomodulin) were increased in those with the ACE DD genotype, indicating that endothelial cell damage is potentiated in those with this genotype.
Angiotensin-converting enzyme genotype and endothelial cell damage in hypertensive patients
von Willebrand factor is a glycoprotein stored in endothelial cells and secreted into the circulation (24), while thrombomodulin is a membrane glycoprotein expressed on the surface of endothelial cells, where it is an important cofactor in the thrombin-catalyzed activation of protein C. Soluble thrombomodulin is also present in human plasma, probably due to proteolysis (25). Although their release mechanisms from endothelial cells are different, in this study both vWF and thrombomodulin showed increased levels in the subjects with the ACE DD genotype. In addition, because vWF is an important component in the aggregation of platelets and their adhesion to subendothelium, increased vWF level per se may promote atherosclerosis and thrombosis (24). Thus, endothelial cell dysfunction or damage is probably potentiated in elderly hypertensive patients with the ACE DD genotype. The increased cardiovascular risk associated with the DD genotype might be mediated through endothelial cell damage.
Relationship between BP level and endothelial cell damage in different ACE genotypes
Increased vWF level was reported in hypertensive patients, especially in those with silent target organ damage (microalbuminuria and silent cerebral infarction) (13,14). In this study, positive correlations between BP and both markers of endothelial cell damage (vWF and thrombomodulin) were found predominantly in those with the II genotype, whereas neither association was significant in those with the DD genotype. Considering that there were no significant differences in BP among the groups with the different ACE genotypes, higher BP might be necessary to potentiate endothelial cell damage in those with the II genotype, while in those with the DD genotype, the ACE D allele itself might be more important than BP level in enhancing endothelial cell damage.
Hypertension-induced potentiation of effect of ACE genotype on endothelial cell damage
In the normotensive subjects, there were no significant differences among the ACE genotypes for the levels of the two markers of endothelial cell damage (vWF and thrombomodulin), suggesting that the effect of ACE D allele on endothelial cell damage is augmented by hypertension. The presence of mild hypertension, in which systolic BP/diastolic BP is 140/90 mm Hg or more, might be sufficient to enhance endothelial cell damage in the hypertensive patients with the DD genotype, whereas in hypertensive patients with the II genotype, higher BP levels are necessary to enhance endothelial cell damage.
Mechanism of relationship between ACE genotype and endothelial cell damage
The mechanism of the relationship between ACE D allele and endothelial cell damage remains uncertain. Compared with homozygous individuals for the I allele, those homozygous for the D allele show a twofold elevation of the plasma level of ACE (26), which is involved in the rate-limiting step for angiotensin II generation in the arterial wall (27,28). Angiotensin II may play crucial roles in endothelial cell damage and focal spasm involving damaged endothelial cells of the coronary artery, since several lines of evidence suggest that it acts as a vasoconstrictor of vascular smooth muscle (29).
Angiotensin-converting enzyme genotype and hypercoagulability in hypertensive patients
In order to assess hypercoagulability, we measured the plasma levels of F1+2. The conversion of prothrombin to thrombin is the central event in the coagulation of blood. This reaction takes place continuously at an appropriate rate under physiological conditions in vivo. During this process, the amino acid terminus of the prothrombin molecule is released as the inactive F1+2 by factor Xa (18). The measurement of this fragment in the bloodstream enables us to assess not only hypercoagulability but also hypocoagulability like the condition seen in warfarin treatment (18,19). Thus, even low-grade hypercoagulability in the atherosclerotic process can accurately be assessed using the level of F1+2. In the present study, no differences in plasma F1+2 levels among the ACE genotype groups were found. Thus, hypercoagulability might not contribute to the cardiovascular risk associated with the ACE DD genotype.
ACE genotype and cardiovascular disease
We recently clarified that the ACE D allele might be associated with ischemic stroke in hypertensive Japanese patients (5). Among asymptomatic elderly Japanese subjects with risk factors (hypertension, dyslipidemia, cigarette smoking or elevated hematocrit), the levels of F1+2, vWF, thrombomodulin and lipoprotein(a) were higher in those with silent multiple lacunar infarction detected by magnetic resonance imaging than in those without any lacunae, indicating that hypercoagulability, endothelial cell damage and high lipoprotein(a) levels contribute to lacunar stroke formation (14). In individuals with these predisposing conditions, the ACE D allele is probably a risk factor for cardiovascular events associated with endothelial cell damage, but it does not seem to be associated with hypercoagulability and high lipoprotein(a).
Racial difference and study limitations
The ACE D allele frequencies of Japanese hypertensive patients (D = 0.38) and normotensive subjects (D = 0.37) were lower than those reported previously for Westerners and for other Japanese populations (D = 0.4–0.6) (1,3,30–33). Because our study population was elderly, there is a possibility of premature death of patients with the DD genotype due to early-onset cardiovascular disease. However, this possibility might be low, because there was no difference in the D allele frequency between younger elderly subjects aged 59 to 75 years and older elderly subjects aged 76 years or more (data not shown). Alternatively, the lower ACE D allele frequency may be characteristic of our district, since our normotensive control group consisted of healthy participants in annual health examinations, who were living in the same area where the hypertensive patients were living. The consistency of our results among different racial populations and the pathological relevance still need to be assessed by larger prospective association studies or linkage-based family studies.
There was considerable overlap in the levels of endothelial cell-derived markers among the patients with ACE II, ID and DD genotypes. We have used the levels of endothelial cell-derived markers to assess the state of the endothelial cells. Although there are no gold standard methods to assess the state of endothelial cells directly, the endothelial cell damage in patients with DD genotype should also be confirmed by other methods to assess the state of the endothelial cells.
Considering the increased plasma levels of both endothelial cell-derived markers in the hypertensive patients with ACE DD genotype, we conclude that the ACE D allele is probably a risk factor for the development of hypertensive cardiovascular disease associated with endothelial cell damage. There is a possibility that advancing age and hypertension might reinforce the positive relationship between the ACE D allele and endothelial cell damage.
☆ This work was supported in part by grants-in-aid from the Foundation for the Development of the Community, Tochigi, Japan.
- angiotensin-converting enzyme
- analysis of variance
- blood pressure
- deoxyribonucleic acid
- left ventricular hypertrophy diagnosed by ECG
- enzyme-linked immunosorbent assay
- prothrombin fragment 1+2
- high density lipoprotein
- polymerase chain reaction
- von Willebrand factor
- Received August 22, 1997.
- Revision received April 8, 1998.
- Accepted April 23, 1998.
- American College of Cardiology
- Raynolds M.V,
- Bristow M.R,
- Bush E.W,
- et al.
- Nakai K,
- Itoh C,
- Miura Y,
- et al.
- Kario K,
- Kanai N,
- Saito K,
- Matsuo T,
- Nago N,
- Shimada K
- Arbustini E,
- Grasso M,
- Fasani R,
- et al.
- Catto A,
- Carter A.M,
- Barrett J.H,
- et al.
- Kario K,
- Kanai N,
- Nishiuma S,
- et al.
- Ludwig E,
- Corneli P.S,
- Anderson J.L,
- Marshall H.W,
- Lalouel J.-M,
- Ward R.H
- Wang X.L,
- McCredie R.M,
- Wilcken D.E.L
- Panahloo A,
- Andres C,
- Mohamed-Ali V,
- et al.
- Kario K,
- Matsuo T,
- Kobayashi H,
- Asada R,
- Matsuo M
- Iwashima Y,
- Sato T,
- Watanabe K,
- et al.
- Sokolow M,
- Werdegar D,
- Kain H.K,
- Hinman A.T
- Kanai N,
- Fujii T,
- Saito K,
- Yokoyama T
- Sadler E.J
- Morishita R,
- Gibbons G.H,
- Kaneda Y,
- Ogihara T,
- Dzau V.J
- Harrap S.B,
- Davidson H.R,
- Connor J.M,
- Soubrier F,
- Corvol P,
- Fraser R