Raloxifene Inhibits Aortic Accumulation of Cholesterol in Ovariectomized, Cholesterol-Fed Rabbits
Background The beneficial effect of long-term hormone replacement therapy in terms of a decreased risk of cardiovascular disease is now generally accepted. Raloxifene, a selective estrogen receptor modulator, has demonstrated hypolipidemic properties while leaving the endometrium unstimulated.
Methods and Results For our study of the effects of raloxifene on atherosclerosis, 75 rabbits were ovariectomized and treated with either raloxifene, 17β-estradiol, or placebo; 25 rabbits were sham operated and treated with placebo. After 45 weeks, the raloxifene group had two thirds of the aortic atherosclerosis, as evaluated by the cholesterol content of the proximal inner part of the aorta, found in the placebo group (placebo, 577±55.1 nmol/mg protein; raloxifene, 397±53.6 nmol/mg protein; P<.05); the estrogen group had one third of the aortic atherosclerosis in the placebo group (estrogen, 177±32.1 nmol/mg protein; P<.001). The sham-operated group (473±59.6 nmol/mg protein) was not significantly different from placebo. These effects were only partly explained by the changes in serum lipids and lipoproteins, and treatment with both estrogen and raloxifene independently predicted the response in aorta cholesterol. Because plasma levels of total raloxifene were low relative to clinical values in postmenopausal women, dose-response data for raloxifene are required.
Conclusions Our findings indicate that raloxifene hydrochloride has a potentially important antiatherogenic effect, analogous to that observed with estrogen in this model.
Cardiovascular disease is a major health problem and the leading cause of death among postmenopausal women in industrialized societies.1 Efficacious preventive strategies and therapies are therefore strongly needed now and in the future. After numerous studies, the protective effect of long-term HRT in CVD is now generally accepted.2 3 However, estrogen monotherapy increases the risk of endometrial cancer, and recent data indicate that long-term estrogen monotherapy and combined estrogen-progestin therapy also elevate the risk of breast cancer.4
Recently, selective estrogen receptor modulators have therefore been introduced in studies for prevention and treatment of postmenopausal osteoporosis. These compounds are characterized by their estrogen-agonistic effects in some tissues (eg, bone) and estrogen-antagonistic effects in other tissues (eg, endometrium). An optimized drug from this category should preferably possess the beneficial properties of estrogen in lack of side effects, ie, endometrial stimulation and the possible increased risk of breast cancer. One of these compounds, raloxifene hydrochloride (C28H27NO4S · HCl), has demonstrated beneficial skeletal effects without endometrial stimulation in both ovariectomized rats5 and postmenopausal women,6 and in vivo rat studies of mammary carcinoma revealed tumor-antagonizing properties.7 In the investigation of these compounds, a crucial question is therefore whether, because of their partial estrogenic activities, they also protect against CVD, as recently indicated by a hypolipidemic effect of raloxifene in rats and postmenopausal women.5 6 Because prospective, randomized, placebo-controlled human studies on the possible cardioprotective effects of these drugs are time-consuming and demand extreme resources, animal studies are interesting. Because the ovariectomized, cholesterol-fed rabbit model of atherosclerosis previously has been shown to be appropriate for studies of estrogen and progestogens, we chose to investigate the effect of raloxifene hydrochloride in this in vivo animal model.
One hundred healthy, sexually mature female rabbits of the Danish Country strain (SSC:CPH) were obtained at Statens Serum Institut, Copenhagen. The rabbits, which previously had been fed a standard commercial rabbit chow, were housed at approved animal facilities (Ledoeje, Denmark) in standard rabbit cages with a room temperature of 20±2°C and a 12-hour light-dark cycle.
The rabbits were allowed to acclimatize for 3 weeks and then were randomly assigned to one of the four treatment groups. In week 4 to 6, they were then ovariectomized (n=75) or sham operated (n=25) during anesthesia with Hypnorm Vet, 0.3 mL/kg IM (0.2 mg fentanyl+10 mg fluanizone/mL, Janssen Pharmaceudica NV).8 To induce atherosclerosis, the rabbits were fed a diet containing 200 mg cholesterol/d from week 8 to 17. During the remaining part of the study (weeks 18 to 45), the dietary amount of cholesterol was reduced to 125 mg chow/d to maintain steady-state levels of serum cholesterol. From week 8 on, the rabbits were orally treated with either 4 mg/d of 17β-estradiol (Novo Nordisk) (25 ovariectomized rabbits), raloxifene 3.5 mg/d (lot No. 255MH3 and 0437D4, Eli Lilly) (25 ovariectomized rabbits), or placebo (25 ovariectomized rabbits and 25 sham-operated rabbits). Because plasma levels of total raloxifene were undetectable during treatment with 3.5 mg/d, the dose was increased to raloxifene 35 mg/d from week 18 to 45. The study was approved by the Danish authorities (Ministry of Justice).
All rabbits were fed 80 g chow/d from week 1 to 17 and 100 g chow/d from week 18 on throughout the study. One portion of chow contained 17β-estradiol (4 mg) dissolved in ethanol (0.25 mL; 100%) with addition of PBS (pH 5.0; 0.1 mL from week 8 to 17, 1 mL from week 18 to 45) or raloxifene (3.5 mg from week 8 to 17, 35 mg from week 18 to 45) suspended in PBS (0.1 mL from week 8 to 17, 1 mL from week 18 to 45) with addition of ethanol (0.25 mL) or placebo (0.25 mL ethanol and 0.1 mL PBS from week 8 to 17, 1 mL PBS from week 18 to 45), 8 mL corn oil (Nomeco, No. BP 80), cholesterol (200 mg from week 8 to 17, 125 mg from week 18 to 45; Struers-Kebolab, No. C-8503), and standard rabbit pellets (72.6 g from week 8 to 17, 90.8 g from week 18 to 45; Brogården, No. 2123). To prepare the chow, the cholesterol and half the corn oil were heated until the cholesterol had completely dissolved and then mixed thoroughly with the rabbit pellets. The solutions containing ethanol and PBS were poured into the remaining corn oil. After evaporation of the ethanol, this was mixed thoroughly into the cholesterol-containing chow. The chow, which was produced for ≈40 days of treatment, was stored in daily portions in darkness at −20°C. The rabbits had free access to water. Food consumption was recorded by collection and weighing of leftover chow weekly. For quality control of the chow, the amount of raloxifene in samples from five batches was determined by HPLC (AAI). Analysis of the raloxifene chow revealed an average recovery of 84.3±2.47% of the originally added amount of raloxifene. For comparison, these analyses were also performed on the same amounts of placebo chow. No raloxifene was recovered from analysis of the placebo chow.
Blood sampling was performed in weeks 4, 14, 23, 31, 39, and 44. Blood samples were drawn 24 hours after the last feed.
Serum total cholesterol, HDL cholesterol (only at baseline), and triglycerides were determined enzymatically by use of Cobas Mira Plus. In addition, serum lipoproteins were fractionated into HDL cholesterol (d>1.063 mg/mL), LDL cholesterol (1.019<d<1.063 mg/mL), IDL cholesterol (1.006<d<1.019 mg/mL), and VLDL cholesterol (d<1.006 mg/mL) by ultracentrifugation. To determine the ultracentrifuged lipoproteins, serum samples were adjusted to the above-mentioned densities by means of solutions of sodium chloride and sodium bromide. After it was sealed, the tube was ultracentrifuged (Beckman L8-70M) at 180×103g/min over 14.5 hours at 4°C (50.4 Ti Beckman rotor). Thereafter, the top and bottom fractions were separated by tube slicing. The cholesterol concentrations of the samples of whole serum and the ultracentrifuged fractions were then determined as described above for total cholesterol. The lipoprotein cholesterol was then determined by calculations after correction for recovery. Ultracentrifugations were performed in weeks 14, 23, 31, 39, and 44.
Plasma Raloxifene Measurements
Plasma total raloxifene was determined in weeks 4, 9, 14, 31, 39, and 44 by HPLC (Eli Lilly).
At the end of the study, the rabbits were euthanized with intravenous injections of a 10% pentobarbital solution. The thoracic aorta was dissected free, and the adventitia was carefully removed under running saline. The aorta was opened longitudinally, and the surface was rinsed with saline. The vessel was then fixed with pins on a corkboard, and the tissue was divided into proximal and distal parts at the level of the first intercostal arteries. From each of these parts, the inner layer containing the intima and part of the media was stripped from the underlying outer media, and the proximal inner part was weighed. These tissues were stored within 1 hour at −20°C until analyzed. The aortic tissue was minced, and the lipids were extracted with chloroform and methanol (2:1, vol/vol) over a period of 24 hours. Lipids and proteins were separated.9 The total cholesterol in the tissue specimens was determined enzymatically after evaporation of the cholesterol-containing fraction and dissolution in isopropyl alcohol.10 The total cholesterol content was adjusted for the amount of protein in the tissue specimens, which was determined by the method of Lowry et al11 after extraction of the lipids and digestion of the residue for 24 hours with 5 mol/L NaOH.
Uterus and Endometrial Tissue
The bicorn uterus of the rabbits was cut 1 cm above the bifurcation and weighed. Thereafter, the uterus was opened, and a sample of endometrial tissue was excised from the cavity, frozen immediately on solid carbon dioxide, and stored at −85°C until analyzed. For the biochemical analysis, the tissue was homogenized (Potter-Elvehjem) and centrifuged at 800g. The 800g supernatant was further centrifuged at 105 000g, and the resulting supernatant (the cytosol) was used for the measurements of cytosolic estrogen binding capacity and cytosolic progesterone binding capacity by steroid-binding assays using dextran-coated charcoal separation as previously described.12 The protein concentration in the cytosol was measured according to Bradford.13 The 800g pellet was washed, nuclear receptors were extracted by 0.6 mol/L KCl,14 and the estradiol receptor content in the nuclear extract was measured by an enzyme immunoassay (Abbott Laboratories). The results of the cytosolic hormone-binding capacities and the nuclear estradiol receptor content were normalized to the cytosolic protein content and expressed as fmol/mg protein. The interassay imprecision of the estradiol-binding capacity, progesterone-binding capacity, estradiol receptor (enzyme immunoassay), and protein measurements were 7%, 10%, 6%, and 5%, respectively.
Body weight was measured on the same scale throughout the study.
All analyses were performed with the Statistical Analyzing System (SAS). The time-averaged concentration of serum lipids and lipoproteins was calculated as the area under the curve divided by the duration of the study period in weeks. One-way ANOVA was used to test for significant differences between the four groups with regard to baseline age, body weight, serum lipids, and lipoproteins. ANOVA was also used to compare food consumption, uterus weight, endometrial activity, and the cholesterol content of the aorta. If the tests revealed significant differences, Student’s t test was used for the group-by-group comparison. The Bonferroni correction did not change the results significantly. A multiple linear regression model (GLM procedure) was used to examine the independent effects of treatment, baseline age and body weight, baseline and time-averaged lipids and lipoproteins, and body weight gain on the aortic accumulation of cholesterol.
Baseline values, weight gain, and food consumption demonstrated comparability of the four groups (Table 1⇓).
From the sham-operated group, 1 rabbit was killed because of recurrent wounds and infections in a paw. In the raloxifene group, 2 rabbits did not complete: 1 was killed because of recurrent cystitis and 1 because of recurrent wounds and infections in a paw. From the estrogen group, 1 rabbit was excluded because it turned out to be male. Thus, 96 rabbits completed the study.
The time-averaged serum lipid and lipoprotein values are presented in Table 2⇓. The time-averaged serum total cholesterol tended to be lower in the raloxifene and estrogen groups than in the two control groups, but not significantly so (Table 2⇓). The time-averaged serum VLDL cholesterol and triglycerides were significantly lower in the estrogen and raloxifene groups than placebo. No differences between the groups were found for LDL cholesterol and IDL cholesterol (Table 2⇓). HDL cholesterol was generally highest in the placebo group; the difference was significant for the estrogen group but not for the raloxifene group.
The cholesterol contents in the inner proximal layer of the thoracic aorta (aortic cholesterol content) in the estrogen and raloxifene groups were ≈33% (177±32.1 nmol/mg protein; P<.001) and ≈66% (397±53.6 nmol/mg protein; P<.05) of the aortic cholesterol content in the placebo group (577±55.1 nmol/mg protein), respectively. The level of aortic cholesterol in the sham-operated group (473±59.6 nmol/mg protein) was slightly lower than placebo, although not significantly so, whereas estrogen-treated rabbits had significantly less aortic content of cholesterol than raloxifene-treated rabbits (P<.01) (Fig 1⇓). Aortic cholesterol content was significantly correlated to the time-averaged concentration of serum total cholesterol (r=.47; P<.001), serum VLDL cholesterol (r=.48; P<.001), serum IDL cholesterol (r=.33; P<.01), and serum triglycerides (r=.34; P<.001) but was not correlated to time-averaged concentrations of LDL and HDL cholesterol. ANCOVA relating aortic cholesterol content as the dependent variable with treatment group and time-averaged serum lipids and lipoproteins as covariates showed that treatment with both estrogen (estimate, −333±67.1 nmol/mg protein; P<.001) and raloxifene (estimate, −152±67.0 nmol/mg protein; P<.05) independently predicted the response in aortic cholesterol content. Correction for baseline serum lipids and lipoproteins, age, and weight did not change this.
Plasma total raloxifene measured in weeks 9 and 14 was undetectable at the initial dose. At the higher dose of 35 mg/d, the average concentrations of plasma total raloxifene in the raloxifene group from weeks 31, 39, and 44 were 47.9±4.7 ng/mL.
Fig 2⇓ illustrates the uterine data. The uterine wet weight was significantly reduced in the ovariectomized compared with the sham-operated rabbits (P<.001). 17β-Estradiol increased uterine wet weight (P<.001), whereas raloxifene did not affect the uterine wet weight. The cytosolic progesterone receptor content in the endometrium was comparably and significantly increased in both the raloxifene (P<.01) and 17β-estradiol groups (P<.001). The group receiving raloxifene had a significantly decreased content of cytosolic estrogen receptors (P<.001), whereas estrogen receptors in the group treated with estrogen tended to be slightly but not significantly lowered. The nuclear content of estrogen receptors in the raloxifene group was increased (P<.001). This was in contrast to the group treated with 17β-estradiol. Ovariectomy influenced neither the content of estrogen nor the content of progesterone receptors.
The principal result from the study was that treatment with raloxifene inhibits aortic cholesterol accumulation in cholesterol-fed ovariectomized rabbits, although not to the same extent as 17β-estradiol therapy. This effect was only partly explained by the changes in serum lipids and lipoproteins, because treatment with both estrogen and raloxifene independently of these variables predicted the response on atherosclerosis.
Previously, results from this rabbit model have been in accordance with data from experiments in baboons and cynomolgus monkeys15 16 and also with findings in humans.2 3 The reliability of our results is also supported by the fact that the magnitude of response to estrogen monotherapy from earlier is nearly identical to that of the present study.17 Only three rabbits died or had to be killed throughout the study, and those remaining seemed healthy, eating the chow and gaining weight. From week 18 on, the amount of chow was increased to 100 g/d, because all rabbits seemed to be rather hungry.
The present study, for the first time, gives an indication of the effect of a selective estrogen receptor modulator, raloxifene, in terms of atherosclerosis. However, at this point uncertainty remains about the maximal effect of raloxifene on atherosclerosis. The response in aortic accumulation of cholesterol to estradiol treatment was approximately twice the response to raloxifene. One reason for this difference may be a question about dosage. The dose of 17β-estradiol was based on previous data to mimic the serum profile in postmenopausal women during HRT.17 It cannot be excluded that our dose regimen of raloxifene might have been too low. Because plasma total raloxifene was undetectable after 2 weeks of low-dose treatment, it was decided to increase the raloxifene dose 10-fold, in turn producing detectable plasma concentrations. Nevertheless, the final level of total plasma raloxifene remained low relative to clinical values in postmenopausal women (Eli Lilly). Conversely, the uterine data confirmed that the present raloxifene dose regimen was biologically active. These data demonstrate that raloxifene exerts both nonestrogenic and estrogenic properties within the same tissue by lacking the uterotrophic effect of estrogen while inducing the expression of one of the proteins known to be synthesized consequent to estrogen stimulation, ie, the progesterone receptor. A dose-response study of raloxifene, however, is necessary to determine the maximal effect of raloxifene on atherosclerosis.
Interestingly, in the raloxifene-treated group, the content of cytosolic estrogen receptors in the endometrium was highly decreased, whereas it was increased in the nuclear compartment. It has been shown that the raloxifene–estrogen-receptor complex, after translocation to the nucleus, associates with regions on DNA specific for the estrogen receptor (estrogen response elements),18 19 analogously to estradiol. However, in contrast to the 17β-estradiol–treated group, the endometrial estrogen receptors from the raloxifene-treated rabbits were retained in the nucleus, an observation described earlier in 4-hydroxytamoxifene–treated mice.20 The reason for and the cellular effect of this translocation are unknown. Generally, the estrogen receptor is considered a nuclear protein, and the receptors measured in the cytosolic fraction constitute the unoccupied receptors. It might be speculated that the much higher proportion of estrogen receptors localized in the nuclear fraction during raloxifene treatment is due to the higher binding affinity between the estrogen receptor and raloxifene compared with the binding affinity between the estrogen receptor and estrogen.21 Another hypothesis may be decreased degradation of the raloxifene–estrogen-receptor complex.
Estrogen receptors have previously been detected in aorta,22 and because raloxifene is thought to act primarily through the estrogen receptor, a hypothesis for the serum lipid–independent effect of both estrogen and raloxifene may involve this receptor. However, the fast responses (20 minutes) seen after intracoronary administration of estrogen suggest that endothelium-dependent but receptor-independent mechanisms also exist. For example, estrogen therapy enhances endothelium-dependent vasodilation in both animals23 and humans,24 and the mechanisms may involve an increase in levels of nitric oxide. Estrogens have, in addition, been shown to influence intra-arterial metabolism of LDL cholesterol,25 perhaps by antioxidative properties. Thus, besides the question of dose-equivalence between raloxifene and 17β-estradiol, the detected difference in response between the two could be attributed to the hypothesis that these treatments might have the same effect on vascular estrogen receptors but that raloxifene potentially lacks the possible non–receptor-mediated antiatherogenic effects of estrogen.
Although not statistically significant, there was a tendency toward a lower aortic cholesterol content in the placebo sham-operated group than in the placebo ovariectomized group. This finding in rabbits is consistent with previous reports in which baseline endogenous estrogen levels do not seem to have significant influence on atherosclerosis compared with ovariectomized rabbits.17 26 The uterine wet weight was significantly lower in the placebo ovariectomized group than in the placebo sham-operated group, whereas there were no differences between rabbits in the placebo ovariectomized and placebo sham-operated groups with regard to endometrial receptor status. These findings suggest a very mild estrogenic effect of endogenous estrogen. It seems that female rabbits are relatively estrogen deficient when left without males. They do not have a regular cycle, and ovulation happens during mating.
An increase in HDL cholesterol, as consistently seen in postmenopausal women during estrogen replacement, was not seen in this study. A partial explanation for this may be found in the enhanced clearance of apolipoprotein E–containing HDL particles via the LDL receptor in rabbits, as suggested by Ma et al.27 Compared with placebo, a neutral effect on HDL cholesterol has been found in ovariectomized, cholesterol-fed rabbits treated with 17β-estradiol, ethinyl estradiol, or 17α-dihydroequilin sulfate.17 26 28 In cholesterol-fed male rabbits treated with intramuscular polyestradiol phosphate and oral ethinyl estradiol, an increase in HDL cholesterol was demonstrated.29 In this study, however, a combination of ultracentrifugation and precipitation was used to determine HDL cholesterol. Thus, methodological differences and the fact that male rabbits were used may explain the difference. In ovariectomized, cholesterol-fed female monkeys, a neutral effect on HDL cholesterol has been found in response to treatment with both 17β-estradiol and conjugated equine estrogens.30 31 Finally, a decrease in HDL cholesterol has been seen in response to a combination of a lipid-lowering diet and conjugated equine estrogens in ovariectomized atherosclerotic monkeys.32 Thus, the response in HDL cholesterol to treatment of cholesterol-fed rabbits and monkeys with estrogens can generally not be extrapolated to that of postmenopausal women during HRT.
Long-term HRT is only sparsely used among postmenopausal women. Approximately 20% of postmenopausal women initiate HRT, and <40% of these women continue the treatment for >1 year.33 The reasons for this are thought to be primarily vaginal bleeding and fear of breast cancer.34 Because HRT needs to be applied long-term to obtain the beneficial effect in preventing osteoporosis and CVD, these potentially adverse effects preclude HRT from exhibiting preventive effects in the postmenopausal population on a larger scale. Our results open up new prospects, because raloxifene not only seems to influence bone while leaving the endometrium and breast unaffected but also, as our data indicate, appears to have a potentially important antiatherosclerotic effect.
Selected Abbreviations and Acronyms
|HPLC||=||high-performance liquid chromatography|
|HRT||=||hormone replacement therapy|
The authors thank the staff at the CCBR animal farm for the daily care of the rabbits and also Eli Lilly, Indianapolis, Ind, for financial support and supplying raloxifene, and Novo Nordisk, Bagsvaerd, Denmark, for supplying 17β-estradiol. In addition, the authors thank B. Potts and Dr M. Knadler for determining plasma levels of raloxifene.
Reprint requests to Dr Nina Hannover Bjarnason, CCBR, Ballerup Byvej 222, DK-2750 Ballerup, Denmark.
- Received January 8, 1997.
- Revision received March 4, 1997.
- Accepted April 12, 1997.
- Copyright © 1997 by American Heart Association
Bush TL, Fried LP, Barret-Connor E. Cholesterol, lipoproteins and coronary heart disease in women. Clin Chem. 1988;34:B60-B70.
Black LJ, Sato M, Rowley ER, Magee DE, Bekele A, Williams DC, Cullinan GJ, Bendele R, Kauffman RF, Bensch WR, Frolik CA, Termine JD, Bryant HU. Raloxifene (LY139481HCl) prevents bone loss and reduces serum cholesterol without causing uterine hypertrophy in ovariectomized rats. J Clin Invest. 1994;93:63-69.
Gottardis MM, Jordan VC. Antitumor actions of keoxifene and tamoxifene in the N-nitrosomethylurea-induced rat mammary carcinoma model. Cancer Res. 1987;47:4020-4024.
Kaplan HM, Timmons EH. The Rabbit: A Model for the Principles of Mammalian Physiology and Surgery. New York, NY: Academic Press Inc; 1979:127-129.
Folch J, Lees M, Stanley GHS. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem. 1957;226:497-509.
Arad Y, Badimon JJ, Badimon L, Hembree WC, Ginsberg HN. Dehydroepiandrosterone feeding prevents aortic fatty streak formation and cholesterol accumulation in cholesterol-fed rabbits. Arteriosclerosis. 1989;9:159-166.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.
Thorpe SM, Lykkesfeldt AE, Vinterby A, Lonsdorfer M. Quantitative immunological detection of estrogen receptors in nuclear pellets from human breast cancer biopsies. Cancer Res. 1986;46:4251s-4255s.
Kushawa RS, Lewis DS, Carey KD, McGill HC. Effects of estrogen and progesterone on plasma lipoproteins and experimental atherosclerosis in the baboon (Papio sp.). Arterioscler Thromb. 1991;11:23-31.
Haarbo J, Leth-Espensen P, Stender S, Christiansen C. Estrogen monotherapy and combined estrogen-progestogen replacement therapy attenuate aortic accumulation of cholesterol in ovariectomized cholesterol-fed rabbits. J Clin Invest. 1991;87:1274-1279.
Yang NN, Venugopalan M, Hardikar S, Glasebrook A. Identification of an estrogen response element activated by metabolites of 17β-estradiol and raloxifene. Science. 1996;273:1222-1225.
Dauvois S, White R, Parker MG. The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling. J Cell Sci. 1993;106:1377-1388.
Bryant HU, Glasebrook AL, Yang NN, Sato M. A pharmacological review of raloxifene. J Bone Miner Metab. 1996;14:1-9.
Lin AL, McGill HC, Shain SA. Hormone receptors of the baboon cardiovascular system. Circ Res. 1982;50:610-616.
Collins P, Rosano GMC, Sarrel PM, Ulrich L, Adamopoulos S, Beale CM, McNeill JG, Poole-Wilson PA. 17β-Estradiol attenuates acetylcholine-induced coronary arterial constriction in women but not men with coronary heart disease. Circulation. 1995;92:24-30.
Wagner JD, Clarkson TB, St Clair RW, Schwenke DC, Shively CA, Adams MR. Estrogen and progesterone replacement therapy reduces low density lipoprotein accumulation in the coronary arteries of surgically postmenopausal cynomolgus monkeys. J Clin Invest. 1991;88:1995-2002.
Sulistiyani, Adelman SJ, Chandrasekaran A, Jayo J, St Clair RW. Effect of 17 alpha-dihydroequilin sulfate, a conjugated equine estrogen, and ethynylestradiol on atherosclerosis in cholesterol-fed rabbits. Arterioscler Thromb Vasc Biol. 1995;15:837-846.
Ma PTS, Yamamoto T, Goldstein JL, Brown MS. Increased mRNA for low density lipoprotein receptor in livers of rabbits treated with 17β-ethinyl estradiol. Proc Natl Acad Sci U S A. 1986;83:792-796.
Adams MR, Kaplan JR, Manuck SB, Korotnik DR, Parks JS, Wolfe MS, Clarkson TB. Inhibition of coronary artery atherosclerosis by 17β-estradiol in ovariectomized monkeys: lack of an effect of added progesterone. Arteriosclerosis. 1990;10:1051-1057.
Williams JK, Honoré EK, Washburn SA, Clarkson TB. Effects of hormone replacement therapy on reactivity of atherosclerotic coronary arteries in cynomolgus monkeys. J Am Coll Cardiol. 1994;7:1757-1761.
Williams JK, Anthony MS, Honoré EK, Herrington DM, Morgan TM, Register TC, Clarkson TB. Regression of atherosclerosis in female monkeys. Arterioscler Thromb Vasc Biol. 1995;15:827-836.