Atherosclerosis and Lipoproteins |
From the Department of Internal Medicine and Biocenter Oulu (A.K., M.J.S., Y.A.K.), University of Oulu, and the Oulu Deaconess Institute (J.H.), Oulu, Finland, and the Orion Corp, Orion Pharma (A.-C.B.), Espoo, Finland.
Correspondence to Prof Y. Antero Kesäniemi, MD, PhD, Department of Internal Medicine, University of Oulu, Kajaanintie 50, 90220 Oulu, Finland. E-mail antero.kesaniemi{at}oulu.fi
| Abstract |
|---|
|
|
|---|
-hydroxylase. In conclusion, the ERT-induced LDL
cholesterollowering effect is related to changes in
estrogen level, which presumably enhance LDL receptor activity, which
is manifested as an increase in FCR for LDL apoB. The small decrease in
the absorption efficiency of dietary cholesterol does not
seem to contribute largely to the cholesterol lowering on
either transdermal or peroral ERT.
Key Words: estrogen replacement therapy LDL cholesterol menopause lipids
| Introduction |
|---|
|
|
|---|
The underlying mechanisms of action of estrogen administered via
various routes are poorly known. The serum estrogen level is varied by
the estrogen regimen and the route of estrogen
administration.8 9 After peroral administration, high
levels of estradiol are catabolized into estrone, which induces protein
synthesis in the liver.10 When the transdermal route is
used, induction of hepatic first-pass metabolism can be
avoided, and a more physiological estrone/estradiol
ratio is achieved.11 Mainly based on animal
studies12 13 or the use of pharmacological doses of
estrogen in prostate cancer,14 the explanation for the LDL
cholesterollowering effect of estrogens has been the
enhanced LDL receptor activity in the liver. The present study was
designed to investigate whether the commonly used doses of estrogen in
replacement therapy might also affect LDL clearance. Potential
additional mechanisms of LDL lowering, such as changes in
cholesterol absorption, were also studied. In addition, the
influence of the polymorphisms of important regulatory proteins,
eg, apoE, apoB, and cholesterol 7
-hydroxylase, on the
lipid response was evaluated. Also, the importance of the route of
administration was elucidated by comparing peroral estradiol valerate
treatment with transdermal estradiol gel therapy.
| Methods |
|---|
|
|
|---|
The criteria for inclusion were as follows: 45 to 65 years of age, a previous hysterectomy with at least 1 remaining ovary, serum follicle-stimulating hormone >30 IU/L, fasting blood glucose <6.7 mmol/L, and BMI <30 kg/m2. Women having contraindications to estrogen therapy or any diseases or medication interfering with lipid metabolism were excluded. The participants were able to keep their lifestyle and concomitant medication during the study.
Height, weight, and waist and hip circumferences were measured with the subjects wearing lightweight clothes without shoes. BMI (weight in kilograms/height in meters squared) and waist-to-hip ratio (waist circumference in centimeters/hip circumference in centimeters) were used to estimate generalized and abdominal obesity, respectively.
Oral and written information was given to the participants, and written informed consent was obtained from all subjects. The study was approved by the ethics committees of the Oulu University and Oulu Deaconess Institute.
Laboratory Analyses
Blood samples were drawn into EDTA-containing tubes in the
morning after an overnight fast. Plasma was separated by
centrifugation at 1200g (2600 rpm) for 15
minutes (4°C). Total plasma cholesterol and
triglyceride levels were determined by enzymatic
colorimetric methods. VLDL, IDL, and LDL were isolated
by repeated ultracentrifugations according to density,
as described in the Manual of Laboratory Operations of the
Lipid Research Clinics Program.16 HDL
cholesterol was determined from VLDL-free plasma after
precipitation of LDLs with heparin-manganese. LDL
cholesterol was also calculated by the Friedewald
formula,17 and these values were used in the LDL response
analyses.
The protein content of lipoproteins was measured by the method of Lowry et al,18 and the amount of apoB was determined after isopropanol precipitation.19 The plasma total apoA I and apoB concentrations were determined by using a commercial kit with a specific selective chemistry analyzer (KONE Instruments Corp).
LDL turnover assessment was carried out as described previously.20 21 In short, 100 mL of fasting blood was drawn for the isolation of LDL, which was carried out according to the method described by Lindgren et al,22 and the LDL protein was labeled with iodine 125 by use of the iodine monochloride method of McFarlane,23 as modified by Bilheimer et al.24 Radiolabeled LDL was injected in the morning on the day after iodination. Blood samples were collected at 0, 15, and 30 minutes and at 1, 2, and 3 hours and thereafter 3 times a week for 14 days after the injection. The radioactivity of total plasma was measured in each sample. The fractional catabolic rate (FCR) was calculated from the plasma decay curves by using the method described by Matthews.25 The production rate of LDL apoB was calculated from FCR, pool volume, and apoB concentration and expressed as milligrams of LDL apoB produced per day normalized for body weight.
Absorption of dietary cholesterol was measured by the peroral double-isotope continuous-feeding method described by Crouse and Grundy.26 Absolute absorption of dietary cholesterol was calculated by multiplying the daily cholesterol intake with the percentage absorption of dietary cholesterol. Seven-day food records were analyzed by a dietitian with the Finnish Food Database Program, Nutrica.27
ApoE phenotype was determined after delipidation with
isoelectric focusing and immunoblotting
techniques28 29 that made use of commercial antibodies.
The EcoRI and XbaI polymorphisms of the apoB
gene and the cholesterol 7
-hydroxylase (CYP7) genes were
determined by polymerase chain reaction as described
previously.30 31
Statistical Analysis
Data analyses were performed with the software packages
SAS (version 6.08) or SPSS for Windows (6.01). The results for
continuous variables are presented as mean±SD. The changes
from baseline to 6 months were analyzed by paired-sample
t test, and the changes between treatments were compared by
independent-sample t test. The effects of estrogen regimens
are presented as mean changes with 95% CI, except the changes
of triglycerides, which are presented as medians
(95% CI). ANOVA with Bonferroni adjustment was used in the group
comparison of apoE phenotypes and the polymorphisms of apoB
and CYP7 genes. In addition, the effect of apoE allele
4 was
studied by comparing the apoE4-negative (including apoE
phenotypes 2/3 and 3/3) with the apoE4-positive (including apoE
phenotypes 4/2, 4/3, and 4/4) subjects. Because of the skewed
distribution of triglycerides, a nonparametric
Mann-Whitney U test and Wilcoxon signed rank test
were used, as appropriate. Spearman correlation coefficients were
calculated to indicate the associations between variables. Stepwise
multiple regression analysis was performed to estimate the
independent factors contributing to the change in LDL
cholesterol and FCR. A value of P<0.05
(2-sided) was considered to indicate statistical significance.
| Results |
|---|
|
|
|---|
|
As reported previously,15 serum estrone was increased more in the peroral group (n=35, from 176±92 to 2176±1156 pmol/L) than in the gel group (n=38, from 224±139 to 418±196 pmol/L; P<0.001 between the treatments). Serum estradiol was increased from 87±133 to 352±213 pmol/L with peroral therapy and from 111±177 to 264±161 pmol/L with the gel therapy (P<0.05 between the treatments).
The LDL cholesterol level decreased by 19%, from 4.19±0.83 (mean±SD) to 3.39±0.78 mmol/L, in the peroral group and by 9%, from 4.11±0.86 to 3.72±0.78 mmol/L, in the gel group. Quite similar decreases of total, VLDL, and IDL cholesterol levels were seen in the study groups, whereas only peroral estrogen increased HDL cholesterol (12%) and total triglycerides (10%) and decreased plasma total and LDL apoB (-12% and -9%, respectively).
Fractional cholesterol absorption was reduced by 10% in
the peroral group (P<0.05) and by 6% in the transdermal
group (P<0.05, Table 2
).
Absolute absorption of dietary cholesterol also decreased
by 18% (P<0.01) and 9% (P<0.05) for peroral
and transdermal therapies, respectively. No correlation was observed
between the changes in the serum estrogen levels and
cholesterol absorption. The change of total and LDL
cholesterol was positively related to the change in
fractional absorption of dietary cholesterol in the peroral
estrogen group (r=0.427, P<0.05 and
r=0.431, P<0.05, respectively), whereas no
significant correlation was observed in the transdermal group
(r=-0.115, P=NS and r=-0.014,
P=NS, respectively).
|
FCR for LDL apoB increased significantly in the peroral estradiol group
(18%), from 0.294 to 0.345 pools per day, but only a minor
insignificant increase (2%) was observed in the transdermal group
(Table 2
). However, the change in LDL cholesterol
was associated with the change in FCR for LDL apoB in both study groups
(Figure 1
). LDL apoB production
was raised by 9% and plasma LDL apoB concentration was lowered by 6%
on the peroral therapy but not on the transdermal gel therapy (Table 2
). The change in FCR for LDL apoB correlated with the change in
serum estrogen (r=0.503, P<0.01 for the change
in estrone; r=0.381, P<0.05 for the change in
estradiol) in the peroral group (Figure 2
), but no significant correlation was
observed in the gel group. However, the change in the
production of LDL apoB was related to the change of serum
estrone on the peroral and transdermal treatments (r=0.380,
P<0.05 and r=0.362, P<0.05,
respectively), whereas no correlation was observed for the change in
serum estradiol levels. The decrease of LDL apoB was related to the
increase in LDL apoB production (r=0.637,
P<0.001) and FCR for LDL apoB (r=-0.571,
P<0.01) for the group on the peroral estrogen replacement
therapy (ERT) compared with the corresponding relations for the group
on transdermal therapy (r=0.634, P<0.001 and
r=-0.241, P=NS, respectively).
|
|
The apoE phenotype distribution was slightly different in the
study groups: there was 1 subject with apoE2/3 in each group, none with
apoE4/2 in the peroral group, and 1 with apoE4/2 in the transdermal
estradiol group, whereas 26 subjects in the peroral and 20 in the
transdermal group had the apoE3/3 phenotype, 5 and 14 subjects
had the apoE4/3 phenotype, and 3 and 2 subjects had the apoE4/4
phenotype, respectively. When the subjects were
analyzed according to their apoE phenotypes,
significant decreases of total and LDL cholesterol were
observed in all apoE phenotypes, and no differences in response
to the treatments were found between the phenotypes. Also, the
changes in other lipids and lipoproteins, LDL turnover, and
cholesterol absorption were quite equal for the different
apoE phenotypes (data not shown). To study whether the effect
of ERT is modified by the
4 allele of apoE, the subjects were
divided into apoE4-negative (phenotypes apoE2/3 and apoE3/3)
and apoE4-positive (phenotypes apoE4/2, apo4/3, and apo4/4)
groups. Although the influence of both estrogen therapies on serum
lipids and cholesterol absorption, FCR, and
production for LDL apoB varied to some extent between the
apoE4-negative and the apoE4-positive subjects, the differences did not
reach statistical significance (data not shown).
No effects of the EcoRI and XbaI
polymorphisms of the apoB gene on the regulation of LDL and
cholesterol metabolism were observed during ERT
(data not shown). Also, no effect of 7
-hydroxylase polymorphism
on lipoprotein or cholesterol metabolism was
observed among the subjects on either treatment (data not shown).
In stepwise multiple regression analysis, the changes in FCR and the production of LDL apoB explained 60% (R2=0.604, P<0.001) of the reduction in LDL cholesterol in the peroral ERT group and 80% (R2=0.798, P<0.001) in the gel group.
| Discussion |
|---|
|
|
|---|
Could the changes in body weight and fat distribution during the study be responsible for the changes in lipoprotein levels? A slight insignificant weight gain (0.6 kg on average) and an increase in BMI with an unchanged waist-to-hip ratio were noticed in the gel group, but these changes can hardly have any adverse effects on lipid metabolism. In the peroral estrogen group, no change in body weight, BMI, or waist-to-hip ratio was observed. In fact, previous studies34 38 have suggested that HRT may prevent weight gain and the tendency of central fat distribution associated with the menopause.39
VLDL cholesterol decreased and VLDL triglycerides tended to decrease on gel therapy, whereas total triglycerides remained unchanged. Walsh et al40 have suggested that the increased production of triglyceride-rich VLDL causes the rise of plasma triglyceride levels on peroral ERT. Similar to findings in the present study, peroral estrogen was shown to increase FCR more than LDL apoB production, resulting in a decrease of plasma LDL levels.40 These metabolic changes were observed in light LDL particles, whereas the clearance and production of dense LDL were increased equally.41
Arca et al42 have suggested that hypercholesterolemia in postmenopausal women is caused by decreased LDL receptor activity. Previously, pharmacological doses of estrogens used in animal studies13 43 or in the treatment of patients with prostate cancer14 44 have also been shown to increase LDL receptor activity. However, there have been only limited data concerning the doses commonly used in replacement therapy. In accord with previous smaller studies,40 41 the present study showed that the serum estrogen levels usually achieved by ERT also decrease LDL cholesterol, mainly by increasing FCR for LDL apoB, and that the effect is related to the change in serum estrogen. Indeed, the large increases in serum estrogen levels on peroral treatment seemed to result in larger reductions in LDL cholesterol levels, and a correlation between the change in serum estradiol and estrone and the change in LDL cholesterol has been noticed.15 The relation between the increase of FCR and the reduction of LDL cholesterol on both treatments indicates that the change of FCR is important. Although the production rate of LDL apoB was increased in the peroral group, the increase of FCR was more profound, leading to a decrease of LDL cholesterol. Overall, the data of the present study suggest that FCR is the most important factor for the lowering of LDL cholesterol independent of the route of administration and at the current therapeutic doses of estrogen. Previous studies with higher doses of estrogen in men and animals have shown that estrogen stimulates hepatic LDL receptor expression,13 14 44 probably mediated by estrogen receptors.45 46 This is also the most likely explanation for the increased FCR seen in the present study with lower doses of estrogen. On the other hand, estrogen-stimulated transcytosis of LDL has recently been suggested to have some importance in cholesterol lowering when desialylated forms of LDL are removed by asialoglycoprotein receptors.47
One additional mechanism for the LDL lowering by estrogen therapy could be the altered cholesterol absorption. A slight but significant decrease in dietary cholesterol absorption was found with both regimens of ERT, suggesting that part of the beneficial effect noticed in lipids and lipoproteins could be mediated by the diminished absorption of cholesterol. It is possible, however, that the overall absorption of intestinal cholesterol (dietary plus biliary) may not have changed. We determined the absorption of dietary cholesterol, which represents about one third of the total intestinal cholesterol pool. At any rate, the changes in cholesterol absorption were not associated with the changes in FCR or the production of LDL apoB, factors that seemed to have a major role in the regulation of LDL cholesterol.
The polymorphisms of some regulatory proteins, such as apoE, apoB,
and 7
-hydroxylase, have been reported to affect the plasma lipid and
lipoprotein levels.29 30 31 48 Also, some studies have
suggested that the response to hypolipidemic therapies could be related
to apoE polymorphism,49 50 even though there are also
contradictory findings.51 52 The present study did not
reveal any such effects of either ERT treatment among these
postmenopausal subjects. Recently, the LDL cholesterol
levels in apoE4-negative subjects were reported to respond more
favorably to HRT than did the levels in apoE4-positive
subjects.53 The differences between that study and our
trial could be due to a number of factors. The study by Heikkinen et
al53 used a combination therapy of estradiol and
cyproterone acetate and introduced a long-term dietary therapy, both of
which might affect the final outcome. Patient selection could also be
important, even though it is unlikely that the hysterectomy in our
study patients could have affected the lipid response. The differences
in the duration of estrogen treatment might also be one factor
explaining the different results. It must also be noticed that because
the 73 subjects of the present study were subdivided by treatment
type and genetic polymorphisms, the numbers in each group became
quite small. Therefore, the power of the present study to find
differences between the apoE, apoB, and 7
-hydroxylase
polymorphisms is probably low.
In conclusion, ERT-induced changes in lipids and lipoproteins are related to the achieved estrogen level on peroral and transdermal therapy. The increase of FCR for LDL apoB seems to be the most important factor in the regulation of the LDL level on both treatments, and although peroral ERT slightly enhanced LDL production, the LDL clearance rate was increased more, with the net effect being a decrease in LDL levels. The small decrease in the absorption of dietary cholesterol does not seem to be important for the reduction in serum cholesterol by peroral and transdermal ERT.
| Acknowledgments |
|---|
-hydroxylase polymorphisms and to Kaisa Ketonen, Saija
Kortetjärvi, Marja-Leena Kytökangas, Tiina Lapinkari,
Anna-Riitta Malinen, Liisa Mannermaa, Eila Saarikoski, and Leena Ukkola
for skillful technical assistance. Received May 25, 1999; accepted September 3, 1999.
| References |
|---|
|
|
|---|
2 macroglobulin in rat
liver. Biochim Biophys Acta. 1997;1359:4858.[Medline]
[Order article via Infotrieve]
This article has been cited by other articles:
![]() |
M. Tanaka, R. Koga, H. Tsuda, K. Imai, S. Abe, T. Masuda, M. Iwamoto, E. Nakazono, T. Kamohara, N. Kinukawa, et al. Subcutaneous Fat Accumulation Shows a Beneficial Correlation with Serum Cholesterol in Postmenopausal Japanese Women Experimental Biology and Medicine, September 1, 2007; 232(8): 1064 - 1070. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Kellokoski, S. M. Poykko, A. H. Karjalainen, O. Ukkola, J. Heikkinen, Y. A. Kesaniemi, and S. Horkko Estrogen Replacement Therapy Increases Plasma Ghrelin Levels J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2954 - 2963. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. R. Pirwany, N. Sattar, I. A. Greer, C. J. Packard, and R. Fleming Supraphysiological concentrations of estradiol in menopausal women given repeated implant therapy do not adversely affect lipid profiles Hum. Reprod., March 1, 2002; 17(3): 825 - 829. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |