| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Laboratory of Atherosclerosis Genetics (T.L., H.J., S.L., T.K.), Department of Clinical Chemistry, Centre for Laboratory Medicine, University Hospital of Tampere and Tampere University Medical School, Department of Clinical Chemistry (T.L.), Tampere; Departments of Diagnostic Radiology (P.D.) and Internal Medicine (S.L.), University Hospital of Tampere, Tampere; Department of Biochemistry (C.E.), National Public Health Institute, Helsinki; and Department of Obstetrics and Gynecology (R.P.) and Internal Medicine (S.L.), University Hospital of Tampere, and University of Tampere (R.P., P.D.), 33521 Tampere, Finland
Address all correspondence and requests for reprints to: Terho Lehtimäki, M.D., Ph.D., Laboratory of Atherosclerosis Genetics, Department of Clinical Chemistry, Center for Laboratory Medicine, University Hospital of Tampere, P.O. Box 2000, 33521 Tampere, Finland. E-mail: . terho.lehtimaki{at}uta.fi
Abstract
Apolipoprotein (apo)E gene 
4 allele carrier status modulates the responses of lipoprotein metabolism to hormone-replacement therapy (HRT). We investigated the effect of long-term HRT on the progression of atherosclerosis in postmenopausal women with or without apoE 4 allele. One hundred forty-one nonsmoking postmenopausal women, 45–71 yr old, were divided into 3 groups based on the use of HRT. The HRT-EVP group (n = 61) used sequential estradiol valerate (EV) plus progestin (P), the HRT-EV group used EV alone (n = 40), and a control group had no HRT. Of these 141 women, 93 participated in a 5-yr follow-up study in 1998. In addition to serum lipid concentration and apoE genotype, the atherosclerosis severity score of the abdominal aorta and carotid arteries was determined by sonography. In apoE4-negative subjects, the progression of atherosclerosis severity score was significantly faster in control than in the HRT groups (genotype-by-time interaction P = 0.0026); whereas in apoE4-positive subjects, there were no significant differences in atherosclerosis severity score progression between the control and HRT groups. The effects of HRT on atherosclerosis progression in subjects with no apo 4-allele seems to be especially beneficial, compared with controls with same phenotype status but without HRT. These results may help us to understand, in more detail, the benefit and possible risk of HRT on atherosclerotic diseases.
ATHEROSCLEROTIC DISEASES ARE the major cause of illness and death in women. The effects of hormone-replacement therapy (HRT) on clinical end-points of atherosclerosis in randomized clinical trials are contradictory to (1, 2, 3) several observational studies that indicate estrogens are beneficial, with respect to the risk of atherosclerosis and coronary heart disease (CHD) in postmenopausal women (4, 5, 6, 7, 8, 9, 10). The mechanisms underlying this discrepancy are unclear, but they are believed to be attributable to selection bias or other unmeasured factors (1). Controversy between results of observational and randomized clinical trials might also have some unknown genetic background (11, 12). A more detailed understanding of this genetic background might help us to understand, in more detail, the benefit and possible risk of HRT use on atherosclerotic diseases. The common polymorphism of apolipoprotein (apo)E gene is associated with the severity of atherosclerosis and an increased mortality caused by CHD (13, 14, 15, 16, 17, 18). Carriers of the 4 allele of the apoE gene have higher risk of atherosclerosis than do noncarriers (13, 14, 15, 16, 17, 18). Moreover, the 4 allele determines the prognosis and the effect on prognosis of simvastatin in survivors of myocardial infarction (12).
The gene for apoE is located on the long arm of chromosome 19 (19q13), where three common alleles [2, 3, and 4 (19, 20)] encode the six major apoE phenotypes. This functional polymorphism, which has several important effects on lipoprotein metabolism, increases serum total and low-density lipoprotein (LDL) cholesterol in the following order: E2/2 < E3/2 < E4/2 < E3/3 < E4/3 < E4/4, in adults (13, 15, 21, 22) and children (22, 23). The apoE genotype influences the risk of angiographically determined CHD (24, 25, 26, 27) and is associated with the severity of atherosclerosis, in autopsy studies (16, 28). The apo 4 allele is also associated with mortality caused by CHD (17, 18, 29). In addition, polymorphism of the apoE gene modulates the responses of lipoprotein metabolism to long-term HRT (30). However, the effect of HRT on the development of atherosclerosis in people with different apoE genotypes has not previously been studied prospectively in postmenopausal women.
We hypothesized that the apoE genotype might modulate the effect of HRT on the development of atherosclerosis. Therefore, we investigated prospectively the effect of long-term HRT on the progression of atherosclerosis in postmenopausal women with or without the apoE4 phenotype, compared with controls of the same apoE4 status.
Materials and Methods
Subjects
In 1993, women attending the private outpatient clinic in Tampere for annual routine gynecological examinations were invited to participate. For the cross-sectional baseline study in 1993 (8), 141 nonsmoking, nondiabetic, postmenopausal women, 45–71 yr old, were enrolled. They had no clinically evident cardiovascular diseases or hypertension and were classified into 3 groups based on the use of HRT. The HRT-EVP group (n = 61) used estradiol valerate (EV) at 2 mg/d for 11 d, followed by EV continued with progestin (P) (levonorgestrel, 0.25 mg/d, n = 40; or medroxyprogesterone acetate, 10 mg/d, n = 21) for 10 d. The HRT-EV group (n = 40) used EV alone, and the control group (n = 40) had never used HRT. In the HRT-EVP and HRT-EV groups, there was a pause of therapy for 7 d after each 21-d cycle. In 1998, all of these 141 women were invited, by letter, to participate in the 5-yr follow-up study; and 93 (67%, 62 in HRT group, 31 controls) of them participated. None of these women discontinued the therapy during follow-up. HRT, when used, was started at the time of menopause, for climacteric symptoms. There are no major differences in baseline characteristics shown in Table 1
, i.e. age, body mass index (BMI), or lipids between dropped controls or HRT-treated subjects, compared with corresponding subjects participating in the follow-up study. In addition, both dropped and participated subjects were nonsmoking, nondiabetic, postmenopausal women without clinically evident cardiovascular diseases, hypertension, or chronic medication.
|
Blood samples
Blood samples for serum lipid and genotype analyses were taken after the subjects had fasted overnight. Sampling took place within 3 wk after the sonography and, for HRT users, during the third week of the hormone regimen. After separation of serum by low-speed centrifugation, the sera were divided into aliquots and stored at -70 C until analyzed.
Sonography
Sonography at baseline and follow-up were performed with Toshiba Sonolayer V SSA 100 equipment (Toshiba Corp., Japan), as reported in detail elsewhere (8). Briefly, all the sonographies were done blinded by one experienced sonographer and radiologist (P.D.). Transverse and longitudinal scans of extracranial carotid arteries were performed bilaterally at four different segments of the carotid (8). Only fibrous and calcified atherosclerotic lesions were considered and were defined as plaques when distinct areas of mineralization or/and focal protrusion into the lumen was identified. The intimal-media far-wall thickness equal to or more than 1.3 mm at any segment in carotid arteries was defined as an atherosclerotic plaque (31), and the total number of plaques (NAP) was calculated. All carotid artery examinations were done with a 5.0-MHz convex transducer probe.
Longitudinal sonographs of the abdominal aorta were obtained at 1-cm intervals, and transverse scans at 2-cm intervals, at the area of three aortic segments. Significant aortic plaques were defined as an intima-media far-wall thickness equal to or more than 3.0 mm (31).
The reproducibility of our sonographic protocol for significant aortic and carotid plaques was also examined; 1 month after the first assessment, 20 randomly selected subjects were invited to a repeated examination. The repeatability of the NAP between the first and second examination was 90% for the carotid artery segment areas and 100% for the aortic segments. The atherosclerosis severity score (ASC) was constructed by dividing the atherosclerosis of the abdominal aorta and carotid arteries into 3 severity classes, i.e. 1 = minor, 2 = moderate, and 3 = severe. The ASC was then calculated as the sum of the severity classes of the aorta and carotid artery. The total NAP was calculated (in cross-sectional study only, because 5 yr data were not available) according to the NAP and severity of plaques (criteria for plaques are given above). The scoring was done blindly, without the knowledge of HRT and E4 status, by one person (P.D.).
Use of apoE phenotyping
For apoE phenotyping, delipidated plasma was used. Analysis used isoelectric focusing, cysteamine treatment, and immunoblotting as described by Menzel and Utermann (32), with minor modifications as described by Lehtimäki et al. (22). The apoE phenotypes were verified by comparing them with known apoE genotype standards; apoE phenotyping was done blindly from coded samples, without the knowledge of arterial sonographic status.
Other laboratory analyses
Lipid measurements were done at baseline and after 5-yr follow-up. Serum total cholesterol and triglycerides were determined by a commercial method (Echtachem 700XR; Eastman Kodak Co., Clinical Products Division, Rochester, NY). Serum HDL cholesterol and its subfractions (HDL2 and HDL3) were separated with a dextran-sulfate-Mg precipitation procedure (33), and the cholesterol content was analyzed with a Monarch 2000 Analyzer (Instrumentation Laboratory, Lexington, KY), using the CHOD-PAP cholesterol reagent (catalog no. 237574; Roche Molecular Biochemicals, Mannheim, Germany) and a primary cholesterol standard (catalog no. 530; Orion Diagnostics, Helsinki, Finland). The LDL cholesterol content was calculated according to the Friedewald formula (34). Quantities of apoA1 and apoB were determined on a Monarch Analyzer by an immunoturbidimetric method (35) (catalog nos. 67265 and 67249, Orion Diagnostics). The interassay coefficient of variation of the total cholesterol assessment was 1.4%; of triglycerides, 1.0%; of HDL cholesterol, 3.7%; of HDL3 cholesterol, 5.2%; of apoA1, 3.8%; and of apoB, 3.5%.
Statistical analyses
One-way ANOVA (ANCOVA), using age and BMI as covariates, was used to assess the statistical differences in lipid and apo values between E4-positive and -negative subjects within the HRT and control groups (see Table 1
). Longitudinal follow-up data were analyzed by ANOVA for repeated measures (RANOVA) to find interactions between treatment groups (controls vs. HRT) and time points. RANOVA was done separately within the apoE4-positive and -negative subject groups. ANCOVA (adjusted for age, BMI, and total cholesterol), with the least-significant-difference post hoc test, was used to test the significance of differences between treatment groups in ASC and total NAP at baseline study. A similar analysis, with the same covariates, was performed for follow-up data. Because of the low number of E4/2 heterozygous (n = 2) and E4/4 (n = 4) homozygous subjects, the E4 allele carriers were combined into one group (E4-positives, E4/3, n = 27), and noncarriers formed another group (E4-negatives), as previously done in other studies (12).
Serum triglycerides were analyzed after logarithmic transformations because of the skewed distribution (the untransformed values are shown). All calculations were done on a computer with the Statistica for Windows, version 5.1 software (Statsoft Inc., Tulsa, OK). Data are presented as mean ± SD unless otherwise stated. A P value less than 0.05 was considered statistically significant.
Results
The baseline distribution of known coronary artery disease risk factors among postmenopausal women, with respect to apoE4 allele status (E4-negative/-positive), are presented in Table 1. In the cross-sectional study, within study groups, the only statistically significant difference between E4-positive and E4-negative subjects occurred in serum total cholesterol in controls. Therefore, serum total cholesterol, in addition to age and BMI, was taken as covariate in further analyses where atherosclerosis severity (Fig. 1, A and B) or total NAP was studied.
|
). After 5 yr of follow-up, the corresponding differences between the HRT-EV and HRT-EVP groups and controls were 35.9% and 35.9% (3.00 vs. 4.68 in controls, P = 0.0009 for both, adjusted for age, BMI, and total cholesterol) (Fig 1C). At baseline, E4-negative subjects on HRT-EV had a mean of 38.8% (2.45 vs. 4.00 in controls) lower total NAP; and E4-negative subjects on HRT-EVP, 40.3% (2.39 vs. 4.00 in controls) smaller total NAP than the controls (ANCOVA, P = 0.007 and P = 0.001, respectively; adjusted for age, BMI, and total cholesterol).
At baseline, the ASC of E4-positive subjects tended to increase in the order: HRT-EV (1.06 ± 0.80), HRT-EVP (1.40 ± 0.63), and controls (1.87 ± 0.83) (ANCOVA for trend, P = 0.021; adjusted for age, BMI, and total cholesterol) (Fig 1B). However, after 5 yr of follow-up, these differences between the two HRT groups and controls disappeared (P = not significant, for all) (Fig 1D). At baseline, E4-positive subjects on HRT-EV had a mean of 50.1% (1.83 vs. 3.67 in controls) lower total NAP; and E4-positive subjects on HRT-EVP, 27.2% (2.67 vs. 3.67 in controls) smaller total NAP than the controls (ANCOVA, P = 0.002 and P = 0.095, respectively; adjusted for baseline age, BMI, and total cholesterol).
Among the E4-negative subjects, the progression rate of ASC differed significantly between HRT users and controls (treatment-group-by-time-point interaction in RANOVA, P = 0.0026, Fig. 2A; adjusted for baseline age, BMI, and mean value of total cholesterol in baseline and follow-up). The corresponding changes in serum lipids during follow-up, according to apoE 4 allele status, are shown in Table 2. There were no statistically significant differences between apoE genotypes in the responses of studied lipid traits (i.e. apoE-genotype-group-by-time-point interactions not significant for all lipids within study groups).
|
|
). The corresponding changes in serum lipids during follow-up, according to apoE 4 allele status, are shown in Table 2. Discussion
These data indicate that the effects of HRT on atherosclerosis progression in subjects with no apo 4-allele seem to be especially beneficial, compared with controls with the same phenotype status but without HRT, regardless of some other cardiovascular risk factors (i.e. age, BMI, and serum cholesterol level). Further, in this study, all subjects were nonsmokers, did not have diabetes, had normal blood pressure, and were otherwise clinically healthy. Also, the time elapsed from menopause was similar in users and nonusers of HRT. In addition, the dietary analysis did not reveal any substantial differences in the use of saturated fat or dietary cholesterol between the HRT groups; and, therefore, our findings are unlikely to be caused by different dietary habits between the HRT and control groups (8). Moreover, background information (see Table 1) demonstrates equality of the study groups (E4-positives vs. E4-negatives) in some important parameters. Because some other important factors that may differ between users and nonusers (i.e. socioeconomic status) were not accounted for, one can say that some unknown factors may have biased our results. Therefore, it would be important to repeat our preliminary observational results in more controlled trials. However, the more favorable effect of HRT on the atherosclerosis progression was seen only in the subgroup of apoE4-negative subjects; whereas in subjects with apoE4-positive genotype, the progression of atherosclerosis tended to be similar in controls and HRT users during follow-up.
Estrogen has never been shown to decrease the occurrence of clinical outcomes, i.e. myocardial infarction or CHD death, in women in randomized clinical trials (1, 3). Despite this, numerous findings from nonrandomized experiments show that postmenopausal estrogen therapy has favorable effects on serum lipoprotein concentrations and slows the development of atherosclerotic diseases (6, 7, 8, 9, 10, 36, 37). Our findings might, at least partly, explain these contradictory results, because, in these previous studies, the possible genotypic effects on HRT response are not considered. The apoE genotype effect may be more pronounced in populations, like Finland, where the apoE4 allele frequency is high and roughly one third of the population are E4-allele carriers, as compared with other populations where apoE4 allele is less frequent (22).
Currently, there are no studies on the combined impact of genetic factors and HRT on atherosclerosis progression. However, there is evidence that apoE polymorphisms may influence the serum cholesterol response to HRT in such a way that apoE4-negative subjects have a more pronounced response than apoE4-positive subjects (30). In line with these results, our study shows that the development of atherosclerosis among apoE4-negative subjects is slower than among apoE4-positive subjects. However, we did not find statistically significant differences in serum total cholesterol responses according to apoE 4 status. It is therefore unlikely that the more favorable effect in apoE4-negative subjects is solely attributable to the greater decrease in serum total or LDL cholesterol during HRT among apoE4-negative subjects than in apoE4-positive subjects (30). Interestingly, in line with our results, Gerdes et al. (12) recently showed that the 4 allele determines the prognosis and the effect on prognosis of simvastatin in survivors of myocardial infarction (12) and that this effect existed after adjusting by serum total cholesterol levels. There is also some evidence that apoE polymorphisms may influence the responses to statins of persons with familial and nonfamilial hypercholesterolemia, apoE4-negative subjects having a more substantial response than apoE4-positive ones (38, 39). On the other hand, the relatively high serum total cholesterol levels among subjects with apoE4 may also respond more favorably to a cholesterol-reducing diet (40), but this may not always be the case (41). Kesäniemi et al. (42) reported that subjects with apoE4/3 and E4/4 phenotypes have more efficient cholesterol absorption than those with other phenotypes. These differences in cholesterol absorption between phenotypes may lead to up-regulation of hepatic LDL-receptors in apoE4-negative subjects and a lowering of serum cholesterol levels. Conversely, efficient uptake of apoE4-containing, triglyceride-rich particles causes hepatic lipid accumulation with down-regulation of LDL-receptors and an increase in serum cholesterol levels. On the other hand, oral estrogen therapy mediates an up-regulation of liver LDL receptors and therefore has an LDL-lowering effect (36). It is possible that up-regulation of LDL receptors, induced by estrogen, is impaired in postmenopausal subjects with the apoE4 allele.
The other possibility is that the apoE genotype modifies the effects of HRT on artery wall cells and thus, atherosclerosis progression. Estrogens reduce the accumulation of cholesterol in the arterial walls of rabbits and monkeys (43, 44); and they protect LDL from oxidation (45), suppress artery wall cell proliferation and migration (46) and synthesis of collagen and elastin in the arterial intima (47). Estrogens also seem to increase endothelium-dependent vasodilatation in healthy postmenopausal women (37, 48). On the other hand, apoE is expressed in different artery wall cells (49, 50, 51, 52, 53); and its receptors (i.e. the LDL receptor, VLDL-receptor, and LDL receptor-related protein) also reside in the arterial wall (54, 55).
Sonographic methods permit the evaluation of atherosclerosis in asymptomatic subjects (56, 57, 58, 59, 60); and hence, the clinical rationale of our asymptomatic atherosclerosis is not clear. However, previous studies have reported that extracoronary arteries are affected by atherosclerosis in parallel with coronary arteries (60) and that atherosclerosis in the abdominal and thoracic aorta seems to be strongly associated with cardiovascular death (61).
In our study, the plaques detected by sonography were mostly uncalcified, constituting less advanced atherosclerotic lesions. Because individuals with more advanced stenotic or occlusive atherosclerotic disease were not included, the association of the apoE genotype with atherosclerosis severity status may have been modified. Even though some previous studies of survivors of myocardial infarction have suggested an association of the 2 allele with atherosclerotic disease (62), the majority of epidemiological studies still favor the presence of an association of the 4 allele with the occurrence of myocardial infarction and angiographically documented coronary artery disease (15, 24, 25, 26).
These data provide evidence for the role of genetic factors in affecting the response of HRT. The longitudinal study indicated that the effect of long-term HRT on the extent of atherosclerosis progression was related to the presence or absence of the apoE4 allele in postmenopausal women. Because our findings are limited to the Finnish population, longitudinal studies from other populations are warranted. However, if our findings are repeated, they may have clinical importance when hormonal treatment of atherosclerotic diseases is planned for postmenopausal women. These observational data support the idea that the effects of HRT on atherosclerosis progression in subjects with no apo 4-allele seem to be especially beneficial, compared with controls with the same phenotype status but without HRT. However, because of its observational nature, this apparent interaction needs confirmation in more controlled and bigger clinical trials. The major point of the study is that the result, if repeated, may help us to understand, in more detail, the benefit and possible risk of HRT use on atherosclerotic diseases.
Acknowledgments
We thank Nina Peltonen for skillful technical assistance.
Footnotes
This work was supported by grants from the Finnish Foundation of Cardiovascular Research (Helsinki, Finland), the Elli and Elvi Oksanen Fund of the Pirkanmaa Fund (under the auspices of the Finnish Cultural Foundation), Emil Aaltonen Foundation, Juho Vainio Foundation, and the Medical Research Fund of the Tampere University Hospital.
Abbreviations: ANCOVA, One-way ANOVA; apo, apolipoprotein; ASC, atherosclerosis severity score; BMI, body mass index; CHD, coronary heart disease; EV, estradiol valerate; HDL, high-density lipoprotein; HRT, hormone-replacement therapy; LDL, low-density lipoprotein; NAP, number of plaques; P, progestin; RANOVA, ANOVA for repeated measures.
Received January 8, 2002.
Accepted June 4, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  P. Ouyang, E. D. Michos, and R. H. Karas Hormone Replacement Therapy and the Cardiovascular System: Lessons Learned and Unanswered Questions J. Am. Coll. Cardiol., May 2, 2006; 47(9): 1741 - 1753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-M. Fan, P. Dastidar, H. Jokela, R. Punnonen, and T. Lehtimaki Hepatic Lipase C-480T Genotype-Dependent Benefit from Long-Term Hormone Replacement Therapy for Atherosclerosis Progression in Postmenopausal Women J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3786 - 3792. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. J. MacLusky Estrogen and Alzheimer's Disease: The Apolipoprotein Connection Endocrinology, July 1, 2004; 145(7): 3062 - 3064. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
