- Email: [email protected]
The effect of oral hormone replacement therapy on lipoprotein profile, resistance of LDL to oxidation and LDL particle size
Maturitas 38 (2001) 287– 295 www.elsevier.com/locate/maturitas
The effect of oral hormone replacement therapy on lipoprotein profile, resistance of LDL to oxidation and LDL particle size Carlos Hermenegildo a, Marı´a Cinta Garcı´a-Martı´nez b, Juan J. Tarı´n c, Angel Lla´cer d, Antonio Cano b,* b
a Research Unit, Hospital Clinico Uni6ersitario de Valencia, E 46010 Valencia, Spain Department of Pediatrics, Obstetrics and Gynecology, Faculty of Medicine and Dentistry, Uni6ersity of Valencia, A6da. Blasco Iban˜ez, 17 E 46010, Valencia, Spain c Department of Animal Biology, Uni6ersity of Valencia, E 46010 Valencia, Spain d Cardiology Department, Hospital Clinico Uni6ersitario de Valencia, E 46010 Valencia, Spain
Received 1 October 2000; accepted 5 December 2000
Abstract Objecti6es: To disclose if oral estradiol (E2), alone or in combination with natural progesterone (P) or medroxyprogesterone acetate (MPA), may modify the oxidizability of low density lipoprotein (LDL), and if the effect is achieved at physiological dosages. LDL oxidizability was assessed by the resistance to oxidation by copper and by the particle size profile, since small particles have increased oxidation susceptibility. Methods: Thirty-three women received two consecutive, two-month length doses of 1 and 2 mg/day of oral E2. They were then randomly assigned to a fourteen-day treatment of 2 mg/day E2 plus either 300 mg/day P or 5 mg/day MPA. A parallel group of experiments was performed on a pool of baseline plasma, where hormones were added at the desired concentration. Lipoprotein levels, resistance of LDL to oxidation, and LDL particle diameter, were measured at baseline and after each treatment. Results: Estradiol reduced LDL levels and increased high density lipoprotein (HDL) and triglycerides. P abolished these changes, whereas MPA only reversed the increase of HDL. Estradiol protected LDL from oxidation in a dose-dependent manner, although only at pharmacological concentrations (1 mM or higher). Both P and MPA were inert at either physiological or pharmacological concentrations. The size of the LDL particles remained unaffected except under MPA, in which it was reduced. Conclusions: Estradiol has a protective effect against LDL oxidation, although only at pharmacological dosages. P and MPA did not limit the E2 action. The size of the LDL particles remained unaltered after each E2 dose, but MPA, and not P, was associated with a diminution. © 2001 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction * Corresponding author. Tel.: + 34-96-3983087; fax: +3496-3864815. E-mail address: [email protected] (A. Cano).
The use of hormone replacement therapy (HRT) has been associated with a decreased inci-
0378-5122/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 5 1 2 2 ( 0 1 ) 0 0 1 7 3 - 6
288
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
dence of cardiovascular disease (CVD) [1]. In HERS, a randomised, placebo-controlled study, the benefits of HRT were only observed after the first year of treatment, thus suggesting that the protection of hormones may work slowly, through reduction of atherosclerosis [2]. Estrogens may act at several steps in the atherogenic process to prevent CVD (for a review, see Ref. [3] and [4]). In addition to inducing a beneficial lipoprotein pattern, some of the benefits of estrogens can be ascribed to their ability to inhibit low density lipoprotein (LDL) oxidation, since oxidative modification of LDL is an important atherogenic factor [5]. The antioxidant effect of estrogens may be investigated by measuring the lag time for the formation of conjugated dienes while LDL is being oxidized by copper [6]. The size profile of the LDL particles is an alternative, indirect approach, since the increased atherogenicity of the smaller particles has been attributed to their increased susceptibility to oxidation [7,8]. Estradiol (E2) has been reported antioxidant potential in in vitro experiments [9,10], but results from clinical studies are controversial, since both protective [11] and inert [12,13] effects have been described. Concerning the LDL size profile, menopausal status has been related with a skewed distribution toward the smaller LDL particles [14,15], thus suggesting that estrogens may regulate their size. However, some investigators have reported that the administration of estrogens to postmenopausal women has been associated with an accumulation of smaller LDL particles [16–19], an observation which, nonetheless, has not been unanimously confirmed [12,20,21]. At present, there is not a clear answer to explain the discrepancies between studies. The possibility that, while moving within the therapeutic dose range, the antioxidant effect of E2 might operate only above a certain threshold of blood concentration, has not been adequately addressed. The use of progestins adds another variable, since progestins may mitigate some beneficial effects of estrogens [22,23]. The type of progestin is of particular interest, natural progesterone (P)
being assigned the most neutral effect on lipoprotein profile [22]. There is insufficient information on the action of progestins on LDL oxidation. While some experimental [10,24] and clinical studies [25] have shown neutral effects, some investigators have found that high dose progestins, and particularly medroxyprogesterone acetate (MPA), oppose the antioxidant action of estrogens [24,26]. The aim of the present investigation was to prospectively explore the effect of oral HRT on LDL oxidation, as measured by the resistance to oxidation by copper and by the size profile of its particles. For that, we designed a protocol involving both in vitro experiments as well as direct assessment of the effects observed in plasma from healthy women under oral HRT. The questions we aimed to answer were, whether there is (1) any detectable effect of oral estrogens on LDL oxidation, (2) a dose-dependent estrogenic effect, (3) an effect of progestins, and (4) a difference between natural P and MPA.
2. Subjects and methods
2.1. Patients Thirty-three postmenopausal women with different hypogonadic symptoms, mainly hot flushes and vaginal dryness, were recruited from the menopause unit of our hospital. Women had reached menopause naturally, had been amenorrheic for at least 1 yr, and had high serum FSH levels (\ 40 IU/l). Subjects were healthy according to their medical history, a basic examination and a routine blood analysis; all were nonsmokers, generally sedentary, and did not take any medication. None of the women had received HRT before the study. A complete gynecologic evaluation was performed before acceptance for the study. This study was approved as complying with the Declaration of Helsinki by the institutional review board at our centre, and written informed consent was obtained from all subjects.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
2.2. Study design The oral sequential regimen consisted of two consecutive periods of two-month length of 1 and 2 mg/day of E2 valerate (Progynova®, Schering Espan˜ a S.A., Madrid, Spain). After the two sequential doses, women were randomised to receive 2 mg of E2 valerate combined with either P or MPA for two additional weeks. Fifteen women received 2 mg of E2 valerate plus 300 mg/day of oral micronized P (Utrogestan®, SEID, Barcelona, Spain) and eighteen women received 2 mg of E2 valerate plus oral MPA (5 mg/day, Progevera®, Upjohn Company, Madrid, Spain). Pills were administered at bedtime. Before the start of treatment and on the last day of each treatment protocol, the women were admitted to the outpatient centre of our hospital after an overnight fast. At the first visit, clinical parameters (height and weight) were assessed and body mass index was calculated.
2.3. Analytical methods 2.3.1. Blood sampling Fasting blood samples were drawn into 2 vacuum tubes. The blood in the first tube was allowed to clot and the serum was separated for the measurement of E2 levels and LDL subclass particle size. The second tube, containing ethylenediaminetetraacetic acid (EDTA), was used to prepare plasma for LDL oxidation assays. EDTA was added to prevent LDL oxidation, with a final concentration of 1 mM. Plasma was prepared from the blood within two hours and was added sucrose to prevent lipoprotein modification, with a final sucrose concentration of 0.6% [27]. Serum and plasma were stored at − 80°C until analysis. A routine analysis of blood samples, including a complete lipid profile, was performed using enzymatic methods by an autoanalyzer AU-5000 Olympus. Serum E2 levels were measured by a specific fluorometric immunoassay (Delfia® Estradiol, EG&G WALLAC, Turku, Finland). The intra- and inter-assay coefficients of variation were less than 10%.
289
2.3.2. LDL oxidation LDL was isolated by sequential ultracentrifugation at density 1.019–1.063 mg/ml using a 50.3 Ti Beckman rotor in a LM-8 Beckman ultracentrifuge (Beckman Instruments, Fullerton, CA). Immediately after isolation, a part of the LDL fraction was desalted by filtration over PD-10 columns (Sephadex G-25M, Pharmacia, Sweden) and eluted with EDTA-free phosphate-buffered saline (PBS). Sample protein content was measured by the BIO-RAD Protein assay method (BIO-RAD Laboratories GmbH, Mu¨ nchen, Germany) using bovine serum albumin as standard, and LDL samples were diluted to a final protein concentration of 50 mg/ml in PBS. The oxidizability of LDL was immediately assessed with a spectrophotometric technique essentially as described by Esterbauer et al. [28]. Oxidation of LDL was started by addition of CuSO4 (final concentration: 10 mM) to 1 ml diluted LDL. Conjugated dienes formation was evaluated by monitoring the change in absorbance at 234 nm every 2 min, during 2 h, at 37°C, using an UVIKON 922 spectrophotometer (Kontron Instruments, Milan, Italy), equipped with a sixposition automatic sample changer. Analyses of the time course of LDL oxidation included measurements of the lag time, the slope of the propagation phase, and maximal light absorbance. The lag time was calculated as the interval between initiation of oxidation (time zero) and the intercept of the tangent of the slope of the absorbance curve during the propagation phase, and represents the phase of the LDL resistance to oxidation before rapid conversion of polyunsaturated fatty acids to conjugated hydroperoxides. To minimise analytical variation, all samples of the same subject were analysed simultaneously. The in vitro experiments were performed essentially in the same way, but plasma from different baseline samples were pooled before LDL isolation, and E2, P and MPA were added to attain the desired concentrations. 2.3.3. LDL particle diameter LDL particle diameter was determined by electrophoresis using the Lipoprint LDL system (Quantimetrix Corporation, Hawthorne, CA), ac-
290
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
cording to manufacturer’s instructions. Briefly, 25 ml of serum and 200 ml of loading gel containing Sudan Black B were added onto the precast gel tube, allowed to photopolymerization and run at 3 mA/tube during 70 min using the Electrophoresis Chamber provided by the manufacturer. After the run, gels were scanned using a Gelprinter Plus System (T.D.I., Alcobendas, Spain) and analysed using the Intelligent Quantifier® v3.0 software (Genomic Solutions, Cambridgeshire, UK). Although seven LDL subfractions can be detected by this method, we employed the Rf (relative mobility) at the highest peak of LDL bands, and the LDL size was calculated using the equation: (LDL size (nm)=(1.429 −Rf) ×25), as reported earlier [29]. According to these authors, LDL subclass pattern was defined as A, when there was a predominance of large LDL particles (LDL size\26.3 nm); pattern B was characterised by a predominance of small LDL particles (LDL sizeB25.8 nm); the remainders were classified as intermediate pattern [29]. The properties of the gel prevented, as described by the manufacturer, any modification in the migration of high density lipoprotein (HDL). The intra- and inter-assay coefficients of variation were lower than 10% [30]. To minimise analytical variations, all samples of the same subject were analysed in the same run.
effects. Levene test was used to test the homogeneity of variance for each dependent variable across all level combinations of the between-subjects factors. P values5 0.05 were considered significant. The statistical analysis was carried out using the statistical package for the social sciences (SPSS Inc., Chicago, IL) v9.0 for Windows.
3. Results
3.1. Clinical data The women age was 49.49 1.4 yr, with an average of 3.89 0.9 yr since the cessation of menses. The body mass index was 25.49 0.8 kg/ m2. Baseline blood analytical data, including glucose, urea, uric acid, creatinine, alkaline phosphatase, aspartate transaminase, glutamicpyruvic transaminase, gamma glutamyl transpeptidase, and lactic dehydrogenase, were normal and remained without significant changes at each of the analytical controls during the study. Blood E2 levels were measured at baseline and after each treatment period (Fig. 1). All blood E2
2.4. Statistical analysis Values shown in the text, tables and figures are means 9 SEM. Repeated-measures ANOVA and Student t-test were applied for comparisons of means. When this test was applied, Box’s M test of the homogeneity of the covariance matrices was used to check whether the dependent variables had the same variance-covariance matrix in each level of the between-subjects factors. Univariate results, which are more likely to detect differences when they exist than in the multivariate approach, were used for testing the withinsubjects factors and the interaction of these factors with the between-subjects factors after accepting Mauchly’s test of sphericity of the variance-covariance matrix. When the sphericity test was rejected, the Greenhouse– Geisser epsilon correction was used for testing the within-subjects
Fig. 1. Serum estradiol levels attained after each treatment step. Serum estradiol levels were measured before (baseline values) and after each oral estradiol treatment with 1 or 2 mg of estradiol (E2) and after combined treatment with 2 mg of E2 plus 300 mg/day of micronized progesterone (P) or 5 mg/day of medroxyprogesterone acetate (MPA). *PB 0.001 vs. baseline values.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295 Table 1 Effects of different doses of oral estradiol, combined or not with progestin, on lipoprotein profiles in postmenopausal womena Treatment
LDL
HDL
Cholesterol
TG
Baseline 1 mg E2 2 mg E2 2 mg E2+P 2 mg E2+MPA
147 9 5 132 97* 138 9 6* 141 911 128 99*
609 2 6893* 649 2* 6594 62 9 4
222 9 6 2139 7 220 9 7 2239 14 20998
789 5 92 9 5* 102 97*,† 759 4** 99 9 4*,‡
a
HDL: high density lipoprotein cholesterol; LDL: low density lipoprotein cholesterol; TG: triglycerides; MPA: medroxyprogesterone acetate. Data are expressed in mg/dl and are means9SEM. *PB0.05 vs. baseline values, †PB0.05 vs. 1 mg of E2-treated group, ‡PB0.001 vs. 2 mg of E2 plus P-treated group and **PB0.05 vs. 1 and 2 mg of E2-treated groups.
values from E2-treated groups (in combination or not with progestins) were significantly different (P B 0.001) from baseline values. Comparison between groups (women treated with P or with MPA) of the blood E2 levels showed no differences at any treatment point. The baseline hypogonadic symptomatology had improved substantially, or disappeared, at the first visit after initiation of treatment.
3.2. Lipid profile Plasma lipoprotein concentrations attained after each treatment are shown in Table 1. Administration of 1 and 2 mg of E2 diminished plasma LDL levels and increased HDL levels (P B 0.05 compared with baseline). P administration reversed the E2-induced effects on both LDL and HDL plasma concentrations. MPA reversed the effects of E2 on HDL levels, whereas did not modify the LDL reduction afforded by E2 treatment. There were no changes in total cholesterol concentration with any treatment. Plasma triglycerides levels were increased significantly, dose-dependent manner, after 1 and 2 mg/day of E2 administration (P B 0.05, when compared to baseline values). P administration in combination with E2 completely reversed triglycerides levels to baseline values, whereas MPA administration did not change their concentration,
291
when compared to the values obtained with 2 mg of E2 alone. There were no differences in between-groups comparisons of lipids levels at any treatment point, with the only exception of different concentration of triglycerides levels after the administration of progestins.
3.3. LDL oxidation Fresh plasma samples from postmenopausal women were used to validate the method. The intra- and inter-assay coefficients of variation were calculated for lag time (2.090.5%, n= 11 and 3.69 0.5%, n= 26, respectively), for the maximal absorbance (3.79 0.9% and 4.29 0.8%, respectively) and the slope of LDL oxidation (4.891.4% and 9.19 1.6%, respectively). The potential influence of freeze-thawing was also analysed, by repeated measures of LDL oxidation parameters in plasma samples stored at − 80°C for up to eight months. There were no substantial differences when results obtained with frozenthawed samples were compared with fresh samples. Table 2 presents the LDL oxidation characteristics. None of the different treatments used in this study changed the LDL oxidation parameters. In addition, there were no differences between groups at any treatment point. To further check that lack of effects, we studied the in vitro addition of E2 on the oxidation of LDL obtained from different pools of plasma. The addition of E2 at physiological range doses (1 and 10 nM) did not increase the lag time to LDL oxidation (data not shown). As shown in Fig. 2, only E2 concentrations above 1 mM (1000 times higher than physiological values) significantly increased the lag time oxidation. The maximum effect was about 218% of control values and was obtained with an E2 concentration of circa 10 mM. To study the effect of P and MPA, we measured the lag time after in vitro treatment with two different concentrations (1 and 10 mM) of each progestin, alone or in combination with 1 mM of E2. There were no changes in the lag time when the progestins were used either alone or in combination with E2.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
292
Table 2 Effects of different doses of oral estradiol, combined or not with progestin, on LDL oxidationa Treatment
Lag time (min)
Maximal absorbance (OD at 234 nm)
Slope (OD234/h)
Baseline 1 mg E2 2 mg E2 2 mg E2+P 2 mg E2+MPA
52.69 1.4 52.79 1.9 50.39 1.5 51.392.8 50.8 93.0
1.129 0.04 1.129 0.05 1.049 0.05 1.1090.11 1.15 90.06
1.69 90.09 1.77 9 0.11 1.54 9 0.10 1.68 90.24 1.72 90.09
a Lag time: interval between initiation of oxidation and the intercept of the tangent of the slope of the absorbance during the propagation phase. Maximal absorbance: maximal absorbance reached by the oxidation of low-density lipoproteins. Slope: slope of the propagation phase. There were not statistical differences between or within groups.
3.4. LDL particle size The analyses of LDL size showed 23 women with pattern A (LDL size\26.3 nm), 5 with a pattern B (LDL sizeB25.8 nm) and 5 with an intermediate pattern at baseline. The average baseline LDL size of the women was 26.489 0.18 nm (Fig. 3). There were no differences when women were treated with 1 and 2 mg/d of E2 in either LDL pattern distribution or in the average LDL particle size. Likewise, combined treatment with P and E2 did not affect the parameters of LDL size. However, MPA administration with E2 induced a significant reduction of the LDL particle size (PB 0.05) when compared with treatment with E2 plus P or with 1 mg of E2 alone.
and MPA, two compounds with contrasting effects on some cardiovascular risk parameters. To our knowledge, our study is the first in which the three parameters, resistance to LDL oxidation, LDL particle size, and lipoprotein profile, are studied together in the same group of postmenopausal women, subjected to increasing doses of oral E2, and after the association of P and MPA with E2. The lipoprotein changes observed in our study, including the contrasting profile between P and MPA, are in general agreement with the well-described effects of oral HRT [22,31,32]. Oxidized LDL is an important risk factor for atherogenesis [33]. The protective effect of estrogens against CVD might be also exerted through impairment
4. Discussion The main objectives of the present work were to assess whether the oral administration of E2 to postmenopausal women determined changes in two parameters that have been implicated in the development of CVD, the lipoprotein profile and the LDL oxidizability. This last parameter was assessed directly, by measurement of the lag time to oxidation by copper and indirectly, by the LDL size profile. Two different oral E2 dosages, representing the usual therapeutic spectrum, were used in order to disclose whether the investigated parameters were only manifest above a certain threshold, or were subjected to a dose-dependent effect. Additionally, the potential effect of progestins was studied through the inclusion of P
Fig. 2. In vitro effect of different concentrations of estradiol on lag time to LDL oxidation. Different pools of LDL were oxidised in vitro in the presence of several estradiol concentrations and the lag time to LDL oxidation measured. Each bar represents mean 9 SEM of 3– 6 pools. *PB0.05 vs. control (0 mM) values.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
Fig. 3. Effects of different doses of oral estradiol, combined or not with progestin, on LDL particle size. LDL particle size (nm) was measured before (baseline values) and after treatment with 1 or 2 mg of oral E2, and after combined treatment with 2 mg of oral E2 with either progesterone (P) or medroxyprogesterone acetate (MPA), as described in Section 2. *PB 0.05 vs. 1 mg of E2- and 2 mg of E2 plus P-treated groups.
of the LDL oxidation process. In agreement with that hypothesis, some clinical studies have found that estrogens augment the lag time of LDL oxidation when administered at therapeutic doses through either the oral [34] or the transdermal route [11,35]. However, some investigators have been unable to confirm those changes [12,25]. The dose-response design of our study, including two different steps of dosage within therapeutic range, confirms the negative studies. We are unable to offer an explanation for the discrepancy. Does it necessarily mean that the hypothesis of an antioxidant effect of estrogens on LDL should be discarded?. In addition to a possible type II error, it is very helpful, at this stage, to look at the in vitro data. Our experiments confirmed that E2 promoted a clear increase of the LDL resistance to oxidation. This effect, however, was only detected at estrogen concentrations which were hundreds of fold higher than physiological levels. To further confirm our data, it is particularly helpful to look at studies performed on premenopausal women, who are subjected to a wide range of endogenous E2 concentrations. Thus, a recent study [36] on women who attained a mean endogenous E2 level of 157 pg/ml was unable to
293
detect any changes in LDL oxidizability. However, other investigators demonstrated an increased resistance to LDL oxidation by copper, but only under the high estrogen levels achieved in women subjected to ovarian hyperstimulation during in vitro fertilisation [37]. The threshold estrogen level that induces protection is not well defined, but the authors only found a positive effect at concentrations above 2000 pg/ml, and, in sharp agreement with the previous study [36], not at any of the estrogen levels found in the distinct phases of a normal menstrual cycle. In our study, the 2 mg E2 dosage achieved a mean E2 concentration of 76.2 pg/ml, which is far from the high values attained in the mentioned study. We cannot discard, nonetheless, that a certain antioxidant effect of estrogens, albeit undetected by our experimental model, might be operative even at the physiological dosages of our study. In fact, one study on hypercholesterolemic menopausal women with coronary heart disease showed that the administration of estrogens did not modify the lag time to LDL oxidation, but decreased the concentration of autoantibodies against oxidized LDL, an indirect indication that estrogens may exert some antioxidant effect [12]. Alternatively, the susceptibility of LDL to oxidation may be examined through the LDL subclass pattern, since small particles are more oxidizable. None of the assayed E2 dosages altered the LDL particle size distribution in our study. This observation is particularly valuable since, as shown in Table 1, the total mass of circulating LDL was decreased after two months under each E2 dose. Consequently, the beneficial reduction of LDL by E2 was not overshadowed by an unfavourable disruption of the LDL distribution size. Our finding agrees with data from other investigators [12,20,21], but is at variance with other reports that describe a reduction of the LDL particles size after estrogen therapy [16– 19,38]. One may speculate that, given the substantial modifications that estrogens impose in the production and catabolism of large and small LDL particles [39], the stable pattern observed in our study further suggests another protective action of estrogens, probably at the level of the LDL subclass production/catabolism equilibrium.
294
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
Progestins have shown different effects in our study. While both P and MPA had neutral effect on LDL susceptibility in in vivo and in vitro studies, they exhibited a contrasting effect on the LDL particle size distribution, since the use of MPA was associated with a diminution. At present, it is unknown which may be the mechanism susceptible to progestin intervention in the LDL size phenotype. However, progestins have consistently demonstrated a counterbalancing effect on the increased triglyceride level associated with estrogen monotherapy, and the level of triglycerides has been shown to keep an inverse correlation with the LDL size distribution pattern [40]. It is revealing that, against natural P, MPA was associated with increased triglyceride levels in our study. Finally, as for LDL resistance to oxidation, it may be argued that the total length of each HRT formulation may be too short, and that some changes may occur at a later term. However, changes in both the lag time to LDL oxidation [11] and in the subclass pattern [18,39] have been detected in shorter time periods. In conclusion, our results indicate that two distinct therapeutic dosages of oral E2 did not modify the lag time to LDL oxidation. This indicator, however, was beneficially modified by the high estrogen dosages employed in the in vitro experiments. The progestins P and MPA were neutral at either physiological or pharmacological dosages. LDL particle size also remained unchanged. The addition of natural P did respect this beneficial effect, but MPA was associated with a significant reduction of the LDL size. This finding may involve increased susceptibility to LDL oxidation.
Acknowledgements Supported by grant 1FD97-1035-C02-01 from the CICYT (Spanish Ministry of Education and Culture) and European Union and by grants GV99-138-1-04 and GV99-6-1-04 from the Consellerı´a de Cultura, Educacio´ i Cie`ncia, Generalitat Valenciana, Spain. The authors are indebted to Mrs. Elvira Calap and Mrs. Rosa Aliaga for their excellent technical assistance.
References [1] Grodstein F, Stampfer MJ, Manson JE, Colditz GA, Willett WC, Rosner B, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med 1996;335:453 – 61. [2] Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605 – 13. [3] Nathan L, Chaudhuri G. Estrogens and atherosclerosis. Annu Rev Pharmacol Toxicol 1997;37:477 – 515. [4] Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N Engl J Med 1999;340:1801 – 11. [5] Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest 1991;88:1785 – 92. [6] Fruebis J, Bird DA, Pattison J, Palinski W. Extent of antioxidant protection of plasma LDL is not a predictor of the antiatherogenic effect of antioxidants. J Lipid Res 1997;38:2455 – 64. [7] Chait A, Brazg RL, Tribble DL, Krauss RM. Susceptibility of small, dense, low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med 1993;94:350 – 6. [8] Galeano NF, Milne R, Marcel YL, Walsh MT, Levy E, Ngu’yen TD, et al. Apoprotein B structure and receptor recognition of triglyceride-rich low density lipoprotein (LDL) is modified in small LDL but not in triglyceriderich LDL of normal size. J Biol Chem 1994;269:511 – 29. [9] McManus J, McEneny J, Young IS, Thompson W. The effect of various oestrogens and progestogens on the susceptibility of low density lipoproteins to oxidation in vitro. Maturitas 1996;25:125 – 31. [10] Arteaga E, Rojas A, Villaseca P, Bianchi M, Arteaga A, Duran D. In vitro effect of estradiol, progesterone, testosterone, and of combined estradiol/progestins on low density lipoprotein (LDL) oxidation in postmenopausal women. Menopause 1998;5:16 – 23. [11] Sack MN, Rader DJ, Cannon RO. Oestrogen and inhibition of oxidation of low-density lipoproteins in postmenopausal women. Lancet 1994;343:269 – 70. [12] Hoogerbrugge N, Zillikens MC, Jansen H, Meeter K, Deckers JW, Birkenhager JC. Estrogen replacement decreases the level of antibodies against oxidized low-density lipoprotein in postmenopausal women with coronary heart disease. Metabolism 1998;47:675 – 80. [13] Wen Y, Doyle MC, Norris LA, Sinnott MM, Cooke T, Harrison RF, et al. Combined oestrogen-progestogen replacement therapy does not inhibit low-density lipoprotein oxidation in postmenopausal women. Br J Clin Pharmacol 1999;47:315 – 21. [14] Campos H, McNamara JR, Wilson PW, Ordovas JM, Schaefer EJ. Differences in low density lipoprotein subfractions and apolipoproteins in premenopausal and post-
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
menopausal women. J Clin Endocrinol Metab 1988;67:30– 5. Ikenoue N, Wakatsuki A, Okatani Y. Small low-density lipoprotein particles in women with natural or surgically induced menopause. Obstet Gynecol 1999;93:566 –70. Rajman I, Lip GY, Cramb R, Maxwell SR, Zarifis J, Beevers DG, et al. Adverse change in low-density lipoprotein subfractions profile with oestrogen-only hormone replacement therapy. QJM 1996;89:771 –8. Tilly-Kiesi M, Lappi M, Puolakka J, Luotola H, Pyorala T, Taskinen MR. Different effects of continuous oestrogen-progestin and transdermal oestrogen with cyclic progestin regimens on low-density lipoprotein subclasses. Eur J Clin Invest 1996;26:1125 –33. Wakatsuki A, Ikenoue N, Sagara Y. Estrogen-induced small low-density lipoprotein particles in postmenopausal women. Obstet Gynecol 1998;91:234 –40. Abbey M, Owen A, Suzakawa M, Roach P, Nestel PJ. Effects of menopause and hormone replacement therapy on plasma lipids, lipoproteins and LDL-receptor activity. Maturitas 1999;33:259 –69. Griffin B, Farish E, Walsh D, Barnes J, Caslake M, Shepherd J, et al. Response of plasma low density lipoprotein subfractions to oestrogen replacement therapy following surgical menopause. Clin Endocrinol (Oxf) 1993;39:463– 8. Roscini AR, Lupattelli G, Siepi D, Pagliaricci S, Pirro M, Mannarino E. Low-density lipoprotein size in primary hypothyroidism. Effects of hormone replacement therapy. Ann Nutr Metab 1999;43:374 –9. The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. JAMA 1995; 273: 199 – 208. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Arterioscler Thromb Vasc Biol 1997;17:217 – 21. Mueck AO, Seeger H, Lippert TH. Estradiol inhibits LDL oxidation: do the progestins medroxyprogesterone acetate and norethisterone acetate influence this effect? Clin Exp Obstet Gynecol 1998;25:26 –8. McManus J, McEneny J, Thompson W, Young IS. The effect of hormone replacement therapy on the oxidation of low density lipoprotein in postmenopausal women. Atherosclerosis 1997;135:73 –81. Zhu X, Bonet B, Knopp RH. Estradiol 17b inhibition of LDL oxidation and endothelial cell cytotoxicity is opposed by progestins to different degrees. Atherosclerosis 2000;148:31 – 41. Kleinveld HA, Hak-Lemmers HLM, Stalenhoef AFH, Demacker PN. Improved measurement of low density lipoprotein susceptibility to copper induced oxidation:
.
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
295
Application of a short procedure for LDL. Clin Chem 1992;38:2066 – 72. Esterbauer H, Striegl G, Puhl H, Rotheneder M. Continuous monitoring of in vitro oxidation of human low density lipoprotein. Free Radic Res Commun 1989;6:67 – 75. Kazumi T, Kawaguchi A, Hozumi T, Nagao M, Iwahashi M, Hayakawa M, et al. Low density lipoprotein particle diameter in young, nonobese, normolipidemic Japanese men. Atherosclerosis 1999;142:113 – 9. Muniz N, Duncan D, Neyer G. A new tool for the automated analysis of LDL subfraction patterns generated by the LipoprintTH LDL system. Paper presented at The Frontiers in Lipoprotein and Vascular, 2000; St. Louis, MO, USA. Cano A, Fernandes H, Serrano S, Mahiques P. Effect of continuous oestradiol-medroxyprogesterone administration on plasma lipids and lipoproteins. Maturitas 1991;13:35 – 42. Chen FP, Lee N, Soong YK. Changes in the lipoprotein profile in postmenopausal women receiving hormone replacement therapy. Effects of natural and synthetic progesterone. J Reprod Med 1998;43:568 – 74. Cox DA, Cohen ML. Effects of oxidized low-density lipoprotein on vascular contraction and relaxation: clinical and pharmacological implications in atherosclerosis. Pharmacol Rev 1996;48:3 – 19. Wilcox JG, Hwang J, Hodis HN, Sevanian A, Stanczyk FZ, Lobo RA. Cardioprotective effects of individual conjugated equine estrogens through their possible modulation of insulin resistance and oxidation of low-density lipoprotein. Fertil Steril 1997;67:57 – 62. Koh KK, Bui MN, Mincemoyer R, Cannon RO. Effects of hormone therapy on inflammatory cell adhesion molecules in postmenopausal healthy women. Am J Cardiol 1997;80:1505 – 7. Wen Y, Doyle MCT, Cooke T, Feely J. Effect of menopause on low-density lipoprotein oxidation: is oestrogen an important determinant? Maturitas 2000;34:233 – 8. Santanam N, Shern-Brewer R, McClatchey R, Castellano PZ, Murphy AA, Voelkel S, et al. Estradiol as an antioxidant: incompatible with its physiological concentrations and function. J Lipid Res 1998;39:2111 – 8. van der Mooren MJ, de Graaf J, Demacker PN, de Haan AF, Rolland R. Changes in the low-density lipoprotein profile during 17 beta-estradiol-dydrogesterone therapy in postmenopausal women. Metabolism 1994;43:799 – 802. Campos H, Walsh BW, Judge H, Sacks FM. Effect of estrogen on very low density lipoprotein and low density lipoprotein subclass metabolism in postmenopausal women. J Clin Endocrinol Metab 1997;82:3955 – 63. Rainwater DL. Lipoprotein correlates of LDL particle size. Atherosclerosis 2000;148:151 – 8.
The effect of oral hormone replacement therapy on lipoprotein profile, resistance of LDL to oxidation and LDL particle size Carlos Hermenegildo a, Marı´a Cinta Garcı´a-Martı´nez b, Juan J. Tarı´n c, Angel Lla´cer d, Antonio Cano b,* b
a Research Unit, Hospital Clinico Uni6ersitario de Valencia, E 46010 Valencia, Spain Department of Pediatrics, Obstetrics and Gynecology, Faculty of Medicine and Dentistry, Uni6ersity of Valencia, A6da. Blasco Iban˜ez, 17 E 46010, Valencia, Spain c Department of Animal Biology, Uni6ersity of Valencia, E 46010 Valencia, Spain d Cardiology Department, Hospital Clinico Uni6ersitario de Valencia, E 46010 Valencia, Spain
Received 1 October 2000; accepted 5 December 2000
Abstract Objecti6es: To disclose if oral estradiol (E2), alone or in combination with natural progesterone (P) or medroxyprogesterone acetate (MPA), may modify the oxidizability of low density lipoprotein (LDL), and if the effect is achieved at physiological dosages. LDL oxidizability was assessed by the resistance to oxidation by copper and by the particle size profile, since small particles have increased oxidation susceptibility. Methods: Thirty-three women received two consecutive, two-month length doses of 1 and 2 mg/day of oral E2. They were then randomly assigned to a fourteen-day treatment of 2 mg/day E2 plus either 300 mg/day P or 5 mg/day MPA. A parallel group of experiments was performed on a pool of baseline plasma, where hormones were added at the desired concentration. Lipoprotein levels, resistance of LDL to oxidation, and LDL particle diameter, were measured at baseline and after each treatment. Results: Estradiol reduced LDL levels and increased high density lipoprotein (HDL) and triglycerides. P abolished these changes, whereas MPA only reversed the increase of HDL. Estradiol protected LDL from oxidation in a dose-dependent manner, although only at pharmacological concentrations (1 mM or higher). Both P and MPA were inert at either physiological or pharmacological concentrations. The size of the LDL particles remained unaffected except under MPA, in which it was reduced. Conclusions: Estradiol has a protective effect against LDL oxidation, although only at pharmacological dosages. P and MPA did not limit the E2 action. The size of the LDL particles remained unaltered after each E2 dose, but MPA, and not P, was associated with a diminution. © 2001 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction * Corresponding author. Tel.: + 34-96-3983087; fax: +3496-3864815. E-mail address: [email protected] (A. Cano).
The use of hormone replacement therapy (HRT) has been associated with a decreased inci-
0378-5122/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 5 1 2 2 ( 0 1 ) 0 0 1 7 3 - 6
288
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
dence of cardiovascular disease (CVD) [1]. In HERS, a randomised, placebo-controlled study, the benefits of HRT were only observed after the first year of treatment, thus suggesting that the protection of hormones may work slowly, through reduction of atherosclerosis [2]. Estrogens may act at several steps in the atherogenic process to prevent CVD (for a review, see Ref. [3] and [4]). In addition to inducing a beneficial lipoprotein pattern, some of the benefits of estrogens can be ascribed to their ability to inhibit low density lipoprotein (LDL) oxidation, since oxidative modification of LDL is an important atherogenic factor [5]. The antioxidant effect of estrogens may be investigated by measuring the lag time for the formation of conjugated dienes while LDL is being oxidized by copper [6]. The size profile of the LDL particles is an alternative, indirect approach, since the increased atherogenicity of the smaller particles has been attributed to their increased susceptibility to oxidation [7,8]. Estradiol (E2) has been reported antioxidant potential in in vitro experiments [9,10], but results from clinical studies are controversial, since both protective [11] and inert [12,13] effects have been described. Concerning the LDL size profile, menopausal status has been related with a skewed distribution toward the smaller LDL particles [14,15], thus suggesting that estrogens may regulate their size. However, some investigators have reported that the administration of estrogens to postmenopausal women has been associated with an accumulation of smaller LDL particles [16–19], an observation which, nonetheless, has not been unanimously confirmed [12,20,21]. At present, there is not a clear answer to explain the discrepancies between studies. The possibility that, while moving within the therapeutic dose range, the antioxidant effect of E2 might operate only above a certain threshold of blood concentration, has not been adequately addressed. The use of progestins adds another variable, since progestins may mitigate some beneficial effects of estrogens [22,23]. The type of progestin is of particular interest, natural progesterone (P)
being assigned the most neutral effect on lipoprotein profile [22]. There is insufficient information on the action of progestins on LDL oxidation. While some experimental [10,24] and clinical studies [25] have shown neutral effects, some investigators have found that high dose progestins, and particularly medroxyprogesterone acetate (MPA), oppose the antioxidant action of estrogens [24,26]. The aim of the present investigation was to prospectively explore the effect of oral HRT on LDL oxidation, as measured by the resistance to oxidation by copper and by the size profile of its particles. For that, we designed a protocol involving both in vitro experiments as well as direct assessment of the effects observed in plasma from healthy women under oral HRT. The questions we aimed to answer were, whether there is (1) any detectable effect of oral estrogens on LDL oxidation, (2) a dose-dependent estrogenic effect, (3) an effect of progestins, and (4) a difference between natural P and MPA.
2. Subjects and methods
2.1. Patients Thirty-three postmenopausal women with different hypogonadic symptoms, mainly hot flushes and vaginal dryness, were recruited from the menopause unit of our hospital. Women had reached menopause naturally, had been amenorrheic for at least 1 yr, and had high serum FSH levels (\ 40 IU/l). Subjects were healthy according to their medical history, a basic examination and a routine blood analysis; all were nonsmokers, generally sedentary, and did not take any medication. None of the women had received HRT before the study. A complete gynecologic evaluation was performed before acceptance for the study. This study was approved as complying with the Declaration of Helsinki by the institutional review board at our centre, and written informed consent was obtained from all subjects.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
2.2. Study design The oral sequential regimen consisted of two consecutive periods of two-month length of 1 and 2 mg/day of E2 valerate (Progynova®, Schering Espan˜ a S.A., Madrid, Spain). After the two sequential doses, women were randomised to receive 2 mg of E2 valerate combined with either P or MPA for two additional weeks. Fifteen women received 2 mg of E2 valerate plus 300 mg/day of oral micronized P (Utrogestan®, SEID, Barcelona, Spain) and eighteen women received 2 mg of E2 valerate plus oral MPA (5 mg/day, Progevera®, Upjohn Company, Madrid, Spain). Pills were administered at bedtime. Before the start of treatment and on the last day of each treatment protocol, the women were admitted to the outpatient centre of our hospital after an overnight fast. At the first visit, clinical parameters (height and weight) were assessed and body mass index was calculated.
2.3. Analytical methods 2.3.1. Blood sampling Fasting blood samples were drawn into 2 vacuum tubes. The blood in the first tube was allowed to clot and the serum was separated for the measurement of E2 levels and LDL subclass particle size. The second tube, containing ethylenediaminetetraacetic acid (EDTA), was used to prepare plasma for LDL oxidation assays. EDTA was added to prevent LDL oxidation, with a final concentration of 1 mM. Plasma was prepared from the blood within two hours and was added sucrose to prevent lipoprotein modification, with a final sucrose concentration of 0.6% [27]. Serum and plasma were stored at − 80°C until analysis. A routine analysis of blood samples, including a complete lipid profile, was performed using enzymatic methods by an autoanalyzer AU-5000 Olympus. Serum E2 levels were measured by a specific fluorometric immunoassay (Delfia® Estradiol, EG&G WALLAC, Turku, Finland). The intra- and inter-assay coefficients of variation were less than 10%.
289
2.3.2. LDL oxidation LDL was isolated by sequential ultracentrifugation at density 1.019–1.063 mg/ml using a 50.3 Ti Beckman rotor in a LM-8 Beckman ultracentrifuge (Beckman Instruments, Fullerton, CA). Immediately after isolation, a part of the LDL fraction was desalted by filtration over PD-10 columns (Sephadex G-25M, Pharmacia, Sweden) and eluted with EDTA-free phosphate-buffered saline (PBS). Sample protein content was measured by the BIO-RAD Protein assay method (BIO-RAD Laboratories GmbH, Mu¨ nchen, Germany) using bovine serum albumin as standard, and LDL samples were diluted to a final protein concentration of 50 mg/ml in PBS. The oxidizability of LDL was immediately assessed with a spectrophotometric technique essentially as described by Esterbauer et al. [28]. Oxidation of LDL was started by addition of CuSO4 (final concentration: 10 mM) to 1 ml diluted LDL. Conjugated dienes formation was evaluated by monitoring the change in absorbance at 234 nm every 2 min, during 2 h, at 37°C, using an UVIKON 922 spectrophotometer (Kontron Instruments, Milan, Italy), equipped with a sixposition automatic sample changer. Analyses of the time course of LDL oxidation included measurements of the lag time, the slope of the propagation phase, and maximal light absorbance. The lag time was calculated as the interval between initiation of oxidation (time zero) and the intercept of the tangent of the slope of the absorbance curve during the propagation phase, and represents the phase of the LDL resistance to oxidation before rapid conversion of polyunsaturated fatty acids to conjugated hydroperoxides. To minimise analytical variation, all samples of the same subject were analysed simultaneously. The in vitro experiments were performed essentially in the same way, but plasma from different baseline samples were pooled before LDL isolation, and E2, P and MPA were added to attain the desired concentrations. 2.3.3. LDL particle diameter LDL particle diameter was determined by electrophoresis using the Lipoprint LDL system (Quantimetrix Corporation, Hawthorne, CA), ac-
290
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
cording to manufacturer’s instructions. Briefly, 25 ml of serum and 200 ml of loading gel containing Sudan Black B were added onto the precast gel tube, allowed to photopolymerization and run at 3 mA/tube during 70 min using the Electrophoresis Chamber provided by the manufacturer. After the run, gels were scanned using a Gelprinter Plus System (T.D.I., Alcobendas, Spain) and analysed using the Intelligent Quantifier® v3.0 software (Genomic Solutions, Cambridgeshire, UK). Although seven LDL subfractions can be detected by this method, we employed the Rf (relative mobility) at the highest peak of LDL bands, and the LDL size was calculated using the equation: (LDL size (nm)=(1.429 −Rf) ×25), as reported earlier [29]. According to these authors, LDL subclass pattern was defined as A, when there was a predominance of large LDL particles (LDL size\26.3 nm); pattern B was characterised by a predominance of small LDL particles (LDL sizeB25.8 nm); the remainders were classified as intermediate pattern [29]. The properties of the gel prevented, as described by the manufacturer, any modification in the migration of high density lipoprotein (HDL). The intra- and inter-assay coefficients of variation were lower than 10% [30]. To minimise analytical variations, all samples of the same subject were analysed in the same run.
effects. Levene test was used to test the homogeneity of variance for each dependent variable across all level combinations of the between-subjects factors. P values5 0.05 were considered significant. The statistical analysis was carried out using the statistical package for the social sciences (SPSS Inc., Chicago, IL) v9.0 for Windows.
3. Results
3.1. Clinical data The women age was 49.49 1.4 yr, with an average of 3.89 0.9 yr since the cessation of menses. The body mass index was 25.49 0.8 kg/ m2. Baseline blood analytical data, including glucose, urea, uric acid, creatinine, alkaline phosphatase, aspartate transaminase, glutamicpyruvic transaminase, gamma glutamyl transpeptidase, and lactic dehydrogenase, were normal and remained without significant changes at each of the analytical controls during the study. Blood E2 levels were measured at baseline and after each treatment period (Fig. 1). All blood E2
2.4. Statistical analysis Values shown in the text, tables and figures are means 9 SEM. Repeated-measures ANOVA and Student t-test were applied for comparisons of means. When this test was applied, Box’s M test of the homogeneity of the covariance matrices was used to check whether the dependent variables had the same variance-covariance matrix in each level of the between-subjects factors. Univariate results, which are more likely to detect differences when they exist than in the multivariate approach, were used for testing the withinsubjects factors and the interaction of these factors with the between-subjects factors after accepting Mauchly’s test of sphericity of the variance-covariance matrix. When the sphericity test was rejected, the Greenhouse– Geisser epsilon correction was used for testing the within-subjects
Fig. 1. Serum estradiol levels attained after each treatment step. Serum estradiol levels were measured before (baseline values) and after each oral estradiol treatment with 1 or 2 mg of estradiol (E2) and after combined treatment with 2 mg of E2 plus 300 mg/day of micronized progesterone (P) or 5 mg/day of medroxyprogesterone acetate (MPA). *PB 0.001 vs. baseline values.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295 Table 1 Effects of different doses of oral estradiol, combined or not with progestin, on lipoprotein profiles in postmenopausal womena Treatment
LDL
HDL
Cholesterol
TG
Baseline 1 mg E2 2 mg E2 2 mg E2+P 2 mg E2+MPA
147 9 5 132 97* 138 9 6* 141 911 128 99*
609 2 6893* 649 2* 6594 62 9 4
222 9 6 2139 7 220 9 7 2239 14 20998
789 5 92 9 5* 102 97*,† 759 4** 99 9 4*,‡
a
HDL: high density lipoprotein cholesterol; LDL: low density lipoprotein cholesterol; TG: triglycerides; MPA: medroxyprogesterone acetate. Data are expressed in mg/dl and are means9SEM. *PB0.05 vs. baseline values, †PB0.05 vs. 1 mg of E2-treated group, ‡PB0.001 vs. 2 mg of E2 plus P-treated group and **PB0.05 vs. 1 and 2 mg of E2-treated groups.
values from E2-treated groups (in combination or not with progestins) were significantly different (P B 0.001) from baseline values. Comparison between groups (women treated with P or with MPA) of the blood E2 levels showed no differences at any treatment point. The baseline hypogonadic symptomatology had improved substantially, or disappeared, at the first visit after initiation of treatment.
3.2. Lipid profile Plasma lipoprotein concentrations attained after each treatment are shown in Table 1. Administration of 1 and 2 mg of E2 diminished plasma LDL levels and increased HDL levels (P B 0.05 compared with baseline). P administration reversed the E2-induced effects on both LDL and HDL plasma concentrations. MPA reversed the effects of E2 on HDL levels, whereas did not modify the LDL reduction afforded by E2 treatment. There were no changes in total cholesterol concentration with any treatment. Plasma triglycerides levels were increased significantly, dose-dependent manner, after 1 and 2 mg/day of E2 administration (P B 0.05, when compared to baseline values). P administration in combination with E2 completely reversed triglycerides levels to baseline values, whereas MPA administration did not change their concentration,
291
when compared to the values obtained with 2 mg of E2 alone. There were no differences in between-groups comparisons of lipids levels at any treatment point, with the only exception of different concentration of triglycerides levels after the administration of progestins.
3.3. LDL oxidation Fresh plasma samples from postmenopausal women were used to validate the method. The intra- and inter-assay coefficients of variation were calculated for lag time (2.090.5%, n= 11 and 3.69 0.5%, n= 26, respectively), for the maximal absorbance (3.79 0.9% and 4.29 0.8%, respectively) and the slope of LDL oxidation (4.891.4% and 9.19 1.6%, respectively). The potential influence of freeze-thawing was also analysed, by repeated measures of LDL oxidation parameters in plasma samples stored at − 80°C for up to eight months. There were no substantial differences when results obtained with frozenthawed samples were compared with fresh samples. Table 2 presents the LDL oxidation characteristics. None of the different treatments used in this study changed the LDL oxidation parameters. In addition, there were no differences between groups at any treatment point. To further check that lack of effects, we studied the in vitro addition of E2 on the oxidation of LDL obtained from different pools of plasma. The addition of E2 at physiological range doses (1 and 10 nM) did not increase the lag time to LDL oxidation (data not shown). As shown in Fig. 2, only E2 concentrations above 1 mM (1000 times higher than physiological values) significantly increased the lag time oxidation. The maximum effect was about 218% of control values and was obtained with an E2 concentration of circa 10 mM. To study the effect of P and MPA, we measured the lag time after in vitro treatment with two different concentrations (1 and 10 mM) of each progestin, alone or in combination with 1 mM of E2. There were no changes in the lag time when the progestins were used either alone or in combination with E2.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
292
Table 2 Effects of different doses of oral estradiol, combined or not with progestin, on LDL oxidationa Treatment
Lag time (min)
Maximal absorbance (OD at 234 nm)
Slope (OD234/h)
Baseline 1 mg E2 2 mg E2 2 mg E2+P 2 mg E2+MPA
52.69 1.4 52.79 1.9 50.39 1.5 51.392.8 50.8 93.0
1.129 0.04 1.129 0.05 1.049 0.05 1.1090.11 1.15 90.06
1.69 90.09 1.77 9 0.11 1.54 9 0.10 1.68 90.24 1.72 90.09
a Lag time: interval between initiation of oxidation and the intercept of the tangent of the slope of the absorbance during the propagation phase. Maximal absorbance: maximal absorbance reached by the oxidation of low-density lipoproteins. Slope: slope of the propagation phase. There were not statistical differences between or within groups.
3.4. LDL particle size The analyses of LDL size showed 23 women with pattern A (LDL size\26.3 nm), 5 with a pattern B (LDL sizeB25.8 nm) and 5 with an intermediate pattern at baseline. The average baseline LDL size of the women was 26.489 0.18 nm (Fig. 3). There were no differences when women were treated with 1 and 2 mg/d of E2 in either LDL pattern distribution or in the average LDL particle size. Likewise, combined treatment with P and E2 did not affect the parameters of LDL size. However, MPA administration with E2 induced a significant reduction of the LDL particle size (PB 0.05) when compared with treatment with E2 plus P or with 1 mg of E2 alone.
and MPA, two compounds with contrasting effects on some cardiovascular risk parameters. To our knowledge, our study is the first in which the three parameters, resistance to LDL oxidation, LDL particle size, and lipoprotein profile, are studied together in the same group of postmenopausal women, subjected to increasing doses of oral E2, and after the association of P and MPA with E2. The lipoprotein changes observed in our study, including the contrasting profile between P and MPA, are in general agreement with the well-described effects of oral HRT [22,31,32]. Oxidized LDL is an important risk factor for atherogenesis [33]. The protective effect of estrogens against CVD might be also exerted through impairment
4. Discussion The main objectives of the present work were to assess whether the oral administration of E2 to postmenopausal women determined changes in two parameters that have been implicated in the development of CVD, the lipoprotein profile and the LDL oxidizability. This last parameter was assessed directly, by measurement of the lag time to oxidation by copper and indirectly, by the LDL size profile. Two different oral E2 dosages, representing the usual therapeutic spectrum, were used in order to disclose whether the investigated parameters were only manifest above a certain threshold, or were subjected to a dose-dependent effect. Additionally, the potential effect of progestins was studied through the inclusion of P
Fig. 2. In vitro effect of different concentrations of estradiol on lag time to LDL oxidation. Different pools of LDL were oxidised in vitro in the presence of several estradiol concentrations and the lag time to LDL oxidation measured. Each bar represents mean 9 SEM of 3– 6 pools. *PB0.05 vs. control (0 mM) values.
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
Fig. 3. Effects of different doses of oral estradiol, combined or not with progestin, on LDL particle size. LDL particle size (nm) was measured before (baseline values) and after treatment with 1 or 2 mg of oral E2, and after combined treatment with 2 mg of oral E2 with either progesterone (P) or medroxyprogesterone acetate (MPA), as described in Section 2. *PB 0.05 vs. 1 mg of E2- and 2 mg of E2 plus P-treated groups.
of the LDL oxidation process. In agreement with that hypothesis, some clinical studies have found that estrogens augment the lag time of LDL oxidation when administered at therapeutic doses through either the oral [34] or the transdermal route [11,35]. However, some investigators have been unable to confirm those changes [12,25]. The dose-response design of our study, including two different steps of dosage within therapeutic range, confirms the negative studies. We are unable to offer an explanation for the discrepancy. Does it necessarily mean that the hypothesis of an antioxidant effect of estrogens on LDL should be discarded?. In addition to a possible type II error, it is very helpful, at this stage, to look at the in vitro data. Our experiments confirmed that E2 promoted a clear increase of the LDL resistance to oxidation. This effect, however, was only detected at estrogen concentrations which were hundreds of fold higher than physiological levels. To further confirm our data, it is particularly helpful to look at studies performed on premenopausal women, who are subjected to a wide range of endogenous E2 concentrations. Thus, a recent study [36] on women who attained a mean endogenous E2 level of 157 pg/ml was unable to
293
detect any changes in LDL oxidizability. However, other investigators demonstrated an increased resistance to LDL oxidation by copper, but only under the high estrogen levels achieved in women subjected to ovarian hyperstimulation during in vitro fertilisation [37]. The threshold estrogen level that induces protection is not well defined, but the authors only found a positive effect at concentrations above 2000 pg/ml, and, in sharp agreement with the previous study [36], not at any of the estrogen levels found in the distinct phases of a normal menstrual cycle. In our study, the 2 mg E2 dosage achieved a mean E2 concentration of 76.2 pg/ml, which is far from the high values attained in the mentioned study. We cannot discard, nonetheless, that a certain antioxidant effect of estrogens, albeit undetected by our experimental model, might be operative even at the physiological dosages of our study. In fact, one study on hypercholesterolemic menopausal women with coronary heart disease showed that the administration of estrogens did not modify the lag time to LDL oxidation, but decreased the concentration of autoantibodies against oxidized LDL, an indirect indication that estrogens may exert some antioxidant effect [12]. Alternatively, the susceptibility of LDL to oxidation may be examined through the LDL subclass pattern, since small particles are more oxidizable. None of the assayed E2 dosages altered the LDL particle size distribution in our study. This observation is particularly valuable since, as shown in Table 1, the total mass of circulating LDL was decreased after two months under each E2 dose. Consequently, the beneficial reduction of LDL by E2 was not overshadowed by an unfavourable disruption of the LDL distribution size. Our finding agrees with data from other investigators [12,20,21], but is at variance with other reports that describe a reduction of the LDL particles size after estrogen therapy [16– 19,38]. One may speculate that, given the substantial modifications that estrogens impose in the production and catabolism of large and small LDL particles [39], the stable pattern observed in our study further suggests another protective action of estrogens, probably at the level of the LDL subclass production/catabolism equilibrium.
294
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
Progestins have shown different effects in our study. While both P and MPA had neutral effect on LDL susceptibility in in vivo and in vitro studies, they exhibited a contrasting effect on the LDL particle size distribution, since the use of MPA was associated with a diminution. At present, it is unknown which may be the mechanism susceptible to progestin intervention in the LDL size phenotype. However, progestins have consistently demonstrated a counterbalancing effect on the increased triglyceride level associated with estrogen monotherapy, and the level of triglycerides has been shown to keep an inverse correlation with the LDL size distribution pattern [40]. It is revealing that, against natural P, MPA was associated with increased triglyceride levels in our study. Finally, as for LDL resistance to oxidation, it may be argued that the total length of each HRT formulation may be too short, and that some changes may occur at a later term. However, changes in both the lag time to LDL oxidation [11] and in the subclass pattern [18,39] have been detected in shorter time periods. In conclusion, our results indicate that two distinct therapeutic dosages of oral E2 did not modify the lag time to LDL oxidation. This indicator, however, was beneficially modified by the high estrogen dosages employed in the in vitro experiments. The progestins P and MPA were neutral at either physiological or pharmacological dosages. LDL particle size also remained unchanged. The addition of natural P did respect this beneficial effect, but MPA was associated with a significant reduction of the LDL size. This finding may involve increased susceptibility to LDL oxidation.
Acknowledgements Supported by grant 1FD97-1035-C02-01 from the CICYT (Spanish Ministry of Education and Culture) and European Union and by grants GV99-138-1-04 and GV99-6-1-04 from the Consellerı´a de Cultura, Educacio´ i Cie`ncia, Generalitat Valenciana, Spain. The authors are indebted to Mrs. Elvira Calap and Mrs. Rosa Aliaga for their excellent technical assistance.
References [1] Grodstein F, Stampfer MJ, Manson JE, Colditz GA, Willett WC, Rosner B, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med 1996;335:453 – 61. [2] Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605 – 13. [3] Nathan L, Chaudhuri G. Estrogens and atherosclerosis. Annu Rev Pharmacol Toxicol 1997;37:477 – 515. [4] Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N Engl J Med 1999;340:1801 – 11. [5] Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest 1991;88:1785 – 92. [6] Fruebis J, Bird DA, Pattison J, Palinski W. Extent of antioxidant protection of plasma LDL is not a predictor of the antiatherogenic effect of antioxidants. J Lipid Res 1997;38:2455 – 64. [7] Chait A, Brazg RL, Tribble DL, Krauss RM. Susceptibility of small, dense, low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med 1993;94:350 – 6. [8] Galeano NF, Milne R, Marcel YL, Walsh MT, Levy E, Ngu’yen TD, et al. Apoprotein B structure and receptor recognition of triglyceride-rich low density lipoprotein (LDL) is modified in small LDL but not in triglyceriderich LDL of normal size. J Biol Chem 1994;269:511 – 29. [9] McManus J, McEneny J, Young IS, Thompson W. The effect of various oestrogens and progestogens on the susceptibility of low density lipoproteins to oxidation in vitro. Maturitas 1996;25:125 – 31. [10] Arteaga E, Rojas A, Villaseca P, Bianchi M, Arteaga A, Duran D. In vitro effect of estradiol, progesterone, testosterone, and of combined estradiol/progestins on low density lipoprotein (LDL) oxidation in postmenopausal women. Menopause 1998;5:16 – 23. [11] Sack MN, Rader DJ, Cannon RO. Oestrogen and inhibition of oxidation of low-density lipoproteins in postmenopausal women. Lancet 1994;343:269 – 70. [12] Hoogerbrugge N, Zillikens MC, Jansen H, Meeter K, Deckers JW, Birkenhager JC. Estrogen replacement decreases the level of antibodies against oxidized low-density lipoprotein in postmenopausal women with coronary heart disease. Metabolism 1998;47:675 – 80. [13] Wen Y, Doyle MC, Norris LA, Sinnott MM, Cooke T, Harrison RF, et al. Combined oestrogen-progestogen replacement therapy does not inhibit low-density lipoprotein oxidation in postmenopausal women. Br J Clin Pharmacol 1999;47:315 – 21. [14] Campos H, McNamara JR, Wilson PW, Ordovas JM, Schaefer EJ. Differences in low density lipoprotein subfractions and apolipoproteins in premenopausal and post-
C. Hermenegildo et al. / Maturitas 38 (2001) 287–295
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
menopausal women. J Clin Endocrinol Metab 1988;67:30– 5. Ikenoue N, Wakatsuki A, Okatani Y. Small low-density lipoprotein particles in women with natural or surgically induced menopause. Obstet Gynecol 1999;93:566 –70. Rajman I, Lip GY, Cramb R, Maxwell SR, Zarifis J, Beevers DG, et al. Adverse change in low-density lipoprotein subfractions profile with oestrogen-only hormone replacement therapy. QJM 1996;89:771 –8. Tilly-Kiesi M, Lappi M, Puolakka J, Luotola H, Pyorala T, Taskinen MR. Different effects of continuous oestrogen-progestin and transdermal oestrogen with cyclic progestin regimens on low-density lipoprotein subclasses. Eur J Clin Invest 1996;26:1125 –33. Wakatsuki A, Ikenoue N, Sagara Y. Estrogen-induced small low-density lipoprotein particles in postmenopausal women. Obstet Gynecol 1998;91:234 –40. Abbey M, Owen A, Suzakawa M, Roach P, Nestel PJ. Effects of menopause and hormone replacement therapy on plasma lipids, lipoproteins and LDL-receptor activity. Maturitas 1999;33:259 –69. Griffin B, Farish E, Walsh D, Barnes J, Caslake M, Shepherd J, et al. Response of plasma low density lipoprotein subfractions to oestrogen replacement therapy following surgical menopause. Clin Endocrinol (Oxf) 1993;39:463– 8. Roscini AR, Lupattelli G, Siepi D, Pagliaricci S, Pirro M, Mannarino E. Low-density lipoprotein size in primary hypothyroidism. Effects of hormone replacement therapy. Ann Nutr Metab 1999;43:374 –9. The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. JAMA 1995; 273: 199 – 208. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Arterioscler Thromb Vasc Biol 1997;17:217 – 21. Mueck AO, Seeger H, Lippert TH. Estradiol inhibits LDL oxidation: do the progestins medroxyprogesterone acetate and norethisterone acetate influence this effect? Clin Exp Obstet Gynecol 1998;25:26 –8. McManus J, McEneny J, Thompson W, Young IS. The effect of hormone replacement therapy on the oxidation of low density lipoprotein in postmenopausal women. Atherosclerosis 1997;135:73 –81. Zhu X, Bonet B, Knopp RH. Estradiol 17b inhibition of LDL oxidation and endothelial cell cytotoxicity is opposed by progestins to different degrees. Atherosclerosis 2000;148:31 – 41. Kleinveld HA, Hak-Lemmers HLM, Stalenhoef AFH, Demacker PN. Improved measurement of low density lipoprotein susceptibility to copper induced oxidation:
.
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
295
Application of a short procedure for LDL. Clin Chem 1992;38:2066 – 72. Esterbauer H, Striegl G, Puhl H, Rotheneder M. Continuous monitoring of in vitro oxidation of human low density lipoprotein. Free Radic Res Commun 1989;6:67 – 75. Kazumi T, Kawaguchi A, Hozumi T, Nagao M, Iwahashi M, Hayakawa M, et al. Low density lipoprotein particle diameter in young, nonobese, normolipidemic Japanese men. Atherosclerosis 1999;142:113 – 9. Muniz N, Duncan D, Neyer G. A new tool for the automated analysis of LDL subfraction patterns generated by the LipoprintTH LDL system. Paper presented at The Frontiers in Lipoprotein and Vascular, 2000; St. Louis, MO, USA. Cano A, Fernandes H, Serrano S, Mahiques P. Effect of continuous oestradiol-medroxyprogesterone administration on plasma lipids and lipoproteins. Maturitas 1991;13:35 – 42. Chen FP, Lee N, Soong YK. Changes in the lipoprotein profile in postmenopausal women receiving hormone replacement therapy. Effects of natural and synthetic progesterone. J Reprod Med 1998;43:568 – 74. Cox DA, Cohen ML. Effects of oxidized low-density lipoprotein on vascular contraction and relaxation: clinical and pharmacological implications in atherosclerosis. Pharmacol Rev 1996;48:3 – 19. Wilcox JG, Hwang J, Hodis HN, Sevanian A, Stanczyk FZ, Lobo RA. Cardioprotective effects of individual conjugated equine estrogens through their possible modulation of insulin resistance and oxidation of low-density lipoprotein. Fertil Steril 1997;67:57 – 62. Koh KK, Bui MN, Mincemoyer R, Cannon RO. Effects of hormone therapy on inflammatory cell adhesion molecules in postmenopausal healthy women. Am J Cardiol 1997;80:1505 – 7. Wen Y, Doyle MCT, Cooke T, Feely J. Effect of menopause on low-density lipoprotein oxidation: is oestrogen an important determinant? Maturitas 2000;34:233 – 8. Santanam N, Shern-Brewer R, McClatchey R, Castellano PZ, Murphy AA, Voelkel S, et al. Estradiol as an antioxidant: incompatible with its physiological concentrations and function. J Lipid Res 1998;39:2111 – 8. van der Mooren MJ, de Graaf J, Demacker PN, de Haan AF, Rolland R. Changes in the low-density lipoprotein profile during 17 beta-estradiol-dydrogesterone therapy in postmenopausal women. Metabolism 1994;43:799 – 802. Campos H, Walsh BW, Judge H, Sacks FM. Effect of estrogen on very low density lipoprotein and low density lipoprotein subclass metabolism in postmenopausal women. J Clin Endocrinol Metab 1997;82:3955 – 63. Rainwater DL. Lipoprotein correlates of LDL particle size. Atherosclerosis 2000;148:151 – 8.