Atherosclerotic Lesion Development in a Novel Ovary-Intact Mouse Model of Perimenopause
http://www.100md.com
动脉硬化血栓血管生物学 2005年第9期
From the Department of Biological Sciences (L.P.M., C.A.D.), Northern Arizona University, Flagstaff; La Jolla Institute for Molecular Medicine (R.L.E., C.L.B.), San Diego, Calif; and the Department of Physiology (P.B.H.), University of Arizona, Tucson.
Reprint requests to Loretta P. Mayer, PhD, Northern Arizona University, Department of Biological Sciences, Box 5640, Flagstaff, AZ 86011-5640. E-mail Loretta.Mayer@nau.edu
Abstract
Objective— Since the unexpected results from the Women’s Health Initiative, the possible protective role of estrogen in preventing heart disease in perimenopausal and postmenopausal women is uncertain. This study examined atherosclerotic lesion development in ovariectomized versus follicle-depleted ovary-intact cholesterol-fed female low-density lipoprotein (LDL) receptor–deficient mice.
Methods and Results— We studied lesion development in LDL receptor–deficient mice that were ovariectomized or follicle depleted with 4-vinylcyclohexene diepoxide (VCD) to induce ovarian failure, then treated ± exogenous 17?-estradiol via pellet implant. At 120 days after start of cholesterol feeding, the extent of lesion in aorta and innominate artery was determined. Lesion area in both locations was similar in vehicle control, VCD-treated, and ovariectomized mice. Replacement with 17?-estradiol caused lesion reduction (P<0.05) in both arterial locations, but it was most efficacious in suppressing innominate lesion area in VCD-treated mice (12.9±5.2%) compared with ovariectomized mice (40.0±6.04%).
Conclusions— Endocrine status associated with the follicle-depleted ovary influences exogenous estradiol effects during the development of atherosclerotic lesions and, in particular, inhibits lesion progression in the innominate artery.
The goal was to determine whether estrogen treatment altered atherosclerosis development in follicle-deplete ovary-intact mice compared with ovariectomized mice. Ovarian contribution of androgen appeared to augment the protective effects of estrogen, suggesting this new model may provide an improved experimental system to analyze atherosclerosis drugs for perimenopausal and postmenopausal women.
Key Words: atherosclerosis ? estrogen supplementation ? perimenopause model
Introduction
Despite women being protected from cardiovascular disease (CVD) until well after menopause, CVD kills more women than men every year in the United States.1 It was presumed for decades that the atheroprotection enjoyed by women was primarily attributable to ovarian estradiol production. Once ovarian production of estradiol stopped because of surgical or spontaneous menopause, protection was lost, and CVD progressed to clinical complications.2,3 On the basis of this presumption, estrogen replacement therapy was initiated during the 1960s and was predicted to be atheroprotective until July 2002, with the announcement of the unexpected findings of increased CVD in the combined hormone therapy arm of the randomized, controlled Women’s Health Initiative (WHI) trial.4
See page 1765
The WHI results contradicted previous observational studies, such as the Nurses’ Health Study, that reported significant protection from CVD with hormone replacement5 but was consistent with the Heart and Estrogen/progestin Replacement Study (HERS) that reported estrogen replacement did not provide protection to women with existing CVD.6 Nonetheless, the WHI results were surprising in providing evidence that combined hormone replacement may be detrimental in healthy women and had a dramatic impact causing many women to stop hormone replacement therapy.7 In February 2004, the estrogen-only arm of WHI was stopped. Interestingly, unlike the findings in women receiving combined estrogen progestin therapy, there was no increase in cardiovascular events in women treated with conjugated equine estrogens alone.8
The putative atheroprotection of estrogen reported in observational human studies is supported by numerous studies demonstrating protection in a variety of animal models of CVD.9,10 Because only human females have extended postmenopausal survival (as much as one third of the life span is postreproductive), there are no appropriate age-dependent animal models of menopause, and studies rely on surgical removal of the ovaries to mimic the endocrine status of postmenopausal women.11 Surgical menopause (total hysterectomy) obviates the perimenopausal transition that can take 10 years and involves multiple ovarian hormonal changes with time.12 Surgical menopause occurs in <13% of postmenopausal women.13 Therefore, the majority of women experience menopause through a natural transition resulting from ovarian aging and follicle depletion and have residual ovarian tissue devoid of estrogen-producing follicles that can produce progesterone and androgen.3
Recently, we reported a method to chemically induce follicle depletion in mice by accelerating the natural process of atresia.14 The industrial chemical 4-vinylcyclohexene diepoxide (VCD) selectively targets primordial and primary ovarian follicles.14 Because larger-growing follicles, the source of ovarian estrogen, are not targeted by VCD, their depletion is only manifest after all nonrenewing primordial follicles are depleted and can no longer serve as a pool for recruitment to larger follicles. Once ovarian estrogen and inhibin production ceases as a result of follicle depletion, the negative feedback to the pituitary is gone, and plasma follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels rise.14,15 In response to elevated LH, cells in the residual ovarian tissue, likely interstitial in origin, continue to produce steroids: progesterone and androgen, primarily.14 Because ovarian follicles are lost gradually and the animal retains residual steroidogenic ovarian tissue, the VCD-treated mouse has an endocrine profile more closely parallel to that in human perimenopause and postmenopause than that in ovariectomized mouse.14
We reported previously that 17?-estradiol replacement resulted in a significant reduction of atherosclerotic lesions in ovariectomized low-density lipoprotein receptor–deficient (LDLR–/–) mice fed a high-fat, high-cholesterol diet, and the effect was unrelated to plasma cholesterol levels.16 The present study was designed to compare the effect of 17?-estradiol treatment on lesion formation in VCD-induced follicle-depleted, ovary-intact mice and ovariectomized mice to determine whether the differences in endocrine milieu altered the extent of lesions formed in the aortic and innominate arteries.
Methods
Animals
C57BL/6J LDLR–/– mice were purchased from The Jackson Laboratory (Bar Harbor, Maine) and bred in the Banka laboratory. All studies were approved by the La Jolla Institute for Molecular Medicine institutional animal care and use committee and comply with National Institutes of Health guidelines. Mice (75 total) were weaned at 28 days (d), initially fed ad libitum a standard mouse chow diet (Diet No. 5015; Harlan Teklad), housed 3 per cage in autoclaved filter-top cages with autoclaved water, and kept on a 12-hour light/dark cycle in a conventional clean room facility. The treatment protocol for VCD induction of follicular depletion has a patent pending (The University of Arizona). VCD-treated mice are commercially available through The Jackson Laboratory (Bar Harbor, Me). As illustrated in Figure 1, sexually immature mice (28 to 35 d old) were randomly selected for each treatment group, weighed, and injected intraperitoneal daily with VCD (Sigma-Aldrich) in sesame oil at 160 mg/kg per day or sesame oil vehicle control for 15 d (n=6 to 13 per treatment group). At sexual maturity (58 d old), bilateral ovariectomy (OVX) was performed on 2 groups of sesame oil–treated mice (n=6 to 9 per group). Mice were anesthetized for surgery using inhalational methoxyflurane (Metofane; Malinckrodt Veterinary) and intraperitoneal ketamine (Ketostat; 100 mg/kg; Ft Dodge Laboratories Inc).
Figure 1. The diagram represents the study as designed to compare endocrine and cardiovascular attributes in normal cycling, VCD-treated, and ovariectomized LDLR–/– mice on a high-fat diet. Treatments and sample collections were performed as indicated.
Female OVX and VCD-treated mice were implanted on day 32 after the onset of VCD treatment with subcutaneous slow-release hormone pellets (Innovative Research of America) designed to release 4 μg per day for 90 d of 17?-estradiol (E2; n=9 to 13). Three days after OVX surgery and day 33 after the onset of VCD treatment, all mice were started on an atherogenic high-cholesterol diet containing 1.25% cholesterol, 6% fat, minimum essential dietary requirements of vitamin E, and no cholate for 12 weeks; day 120 (Diet No. TD96335; Harlan Teklad).
Tissues
Heparinized venous blood was collected from the retro-orbital sinus under anesthesia after an 8-hour fast unless otherwise specified. Blood samples were collected from animals on a chow diet to correspond to animals on high-fat diet collected at 25, 53, 81, 109, and 120 d after the onset of VCD treatment (Figure 1). The only nonfasting samples were collected at killing (90 d after surgery and 120 d after the onset of VCD treatment). Plasma was obtained by centrifugation and stored (–20°C) before analysis. Mice were weighed while anesthetized. A cumulative plasma pool was generated for steroid hormone measurements by pooling 125 μL of plasma from each blood sample collection time point on day 25, 53, 81, 109, and 120 after the onset of VCD treatment.
On day 120 after VCD treatment, animals were weighed, bled, and perfused under anesthesia. An incision was made in the inferior vena cava, and a cannula was inserted into the left ventricle for perfusion with PBS followed by formal sucrose (4% paraformaldehyde and 5% sucrose in PBS; Sigma-Aldrich). The adrenals, spleen, uteri, livers, kidneys, and ovaries were collected.
All organs were trimmed free of fat and wet weights recorded. All organ weights were normalized to body weight to correct for differences in total body weight. Uteri were excised from the oviduct at the uterotubal junction bilaterally and transected at the junction of the uterine body and cervix. After removal of luminal fluid, the entire uterus was weighed on a microbalance. One ovary from each treatment group was prepared for histological evaluation (n=6 animals per group). The ovary was placed in 4% paraformaldehyde fixative, transferred to 70% ethanol, paraffin embedded and serially sectioned (4 to 5 μm), mounted, and stained with hematoxylin and eosin. Follicles were counted in every twentieth section and classified as described previously.14
Innominate arteries were isolated, cleaned, and photographed in situ under the dissecting microscope using an Olympus DP11 digital camera. Photographic images were subsequently analyzed morphometrically by drawing a line across the origin from the aorta and a parallel line intersecting the bifurcation to the subclavian and common carotid arteries (Figure 2, red lines). This defined the total area of the innominate artery. Lesion area was outlined and the percentage of the total area covered by lesion calculated using NIH Image software. Using a dissecting microscope, the entire mouse aorta was dissected, removed, and prepared as described previously.16 The Sudan IV–stained aortae were photographed at a fixed magnification using an Olympus DP11 and imported into a computer for morphometric analysis (NIH Image software) as described previously.16
Figure 2. Immature female LDLR–/– mice were treated daily with vehicle control, VCD (160 mg/kg per day IP for 15 d; n=6 to 12), or were ovariectomized and received 17?-estradiol supplementation as indicated. Innominate arteries (A) and aortae (B) were collected and prepared on day 120 after onset of dosing (*P<0.05 different from control; letters different within group).
Biochemical Analyses
Hepatocellular histopathology and enzymatic activity of circulating aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were determined by the diagnostic laboratory at the Arizona Health Sciences Center. Total cholesterol and triglycerides were quantified using Sigma reagents 401-25P and 344-20, respectively. High-density lipoprotein (HDL) cholesterol was measured using the Sigma 401-25P reagent after precipitation of the apolipoprotein B–containing lipoproteins (reagent 543004; Boehringer Mannheim).
Plasma levels of 17?-estradiol and androstenedione were determined by Coat-A-Count radioimmunossay (RIA; Diagnostic Products Inc.). Sensitivity and interassay and intra-assay coefficients of variation were 17?-estradiol, 1.4 pg/mL, 4.6% and 14%; androstenedione, 0.04 ng/mL, 8.4% and 5.7%, respectively. Plasma FSH was measured by RIA according to instructions with the murine kit from the National Hormone and Pituitary Distribution Program. All samples were assayed in duplicate. Sensitivity and interassay and intra-assay coefficients of variation were 200 pg/mL, 2.7%, and 6.7%. The results of all RIAs were calculated by 4-parameter logistic analysis using the software Assay Zap (BioSoft).
Data Analysis
Body weights, tissue weights, and plasma hormone concentrations were determined from individual animals and averaged for each treatment, and the means (±SEM) in control versus treated animals were analyzed for significant differences by 1-way ANOVA with significance set at P<0.05. Post hoc tests were used where appropriate, with Dunnett’s used to determine differences from control and Student t test to compare each pair. Tests for homogeneity of variance (Bartlett) and normality (Shapiro–Wilk) were routinely performed to ensure that the assumptions of the ANOVA were met. Lipid measurements were analyzed using a Friedman repeated-measures ANOVA on ranks. The Student-Newman–Keuls method was used for pairwise multiple comparisons of groups as a post hoc test.
Results
Innominate Atherosclerotic Lesion Area
The extent of innominate lesion area was similar and not significantly different among control (47.3±4.8%), VCD (45.04±5.7%), and OVX (57.0±7.3%) treatment groups ranging from 40% to 60% (Figure 2A). However, E2 supplementation in VCD-treated mice resulted in significant differences in lesion area compared with that in OVX mice. The VCD+E2 (12.9±5.2%) treatment group had less (P<0.05) lesion area than the OVX+E2 (40.0±6.0%) treatment group (n=6 to 12).
Aortic Atherosclerotic Lesion Area
The extent of aortic lesion area was similar and not significantly different among the control (6.4±0.7%), VCD (6.8±0.7%), and OVX (8.3±1.5%) groups, ranging from 5% to 8% (Figure 2B). In VCD+E2 (1.3±0.9%) and OVX+E2 (1.7±0.7%), there was a reduction (P<0.05) in aortic lesion area; however, the extent of E2 effects in these 2 treatment groups was not different (n=6 to 12).
Systemic and Liver Function
Body weights at day 120 did not differ significantly among the 5 groups: vehicle, VCD, VCD+17?-estradiol (E2), OVX, OVX+E2 (data not shown). There was no effect of either VCD±E2 or OVX±E2 on adrenal, kidney, or spleen weights normalized to body weight (data not shown). Levels of liver enzymes AST and ALT were within the normal range for VCD-treated and control animals (AST; control, 157.85±56.55 U/L; VCD, 57.30±2.30 U/L: normal range 54 to 298 U/L: ALT; control, 34.53±10.31 U/L; VCD, 35.84±12.40 U/L: normal range 17 to 77 U/L). ALT and AST circulating levels were not analyzed in OVX±E2 mice. There was no histological evidence of hepatocellular pathology in livers from VCD±E2 or OVX±E2 mice (data not shown).
Lipids
Over the time course of the experiment, once fat feeding had commenced, plasma total cholesterol increased 3- to 4-fold from day 25 to day 53 in all experimental groups: vehicle, VCD, VCD+E2, OVX, and OVX+E2 (Figure 3). On day 53, total cholesterol was significantly less in VCD+E2, OVX, and OVX+E2 compared with vehicle control mice. On day 81, plasma total cholesterol was elevated in VCD and OVX mice compared with vehicle control. At day 120, OVX mice had significantly elevated cholesterol compared with the other treatment groups, and there was a significant reduction in plasma cholesterol in OVX+E2 mice. On day 120, neither HDL nor triglycerides differed among any of the groups, suggesting that the significant decrease observed in plasma total cholesterol in OVX+E2 was attributable to reduction in LDL cholesterol because there were no significant reductions in plasma triglycerides, indicating that the affected fraction was LDL and not very–low-density lipoprotein (Table 1).
Figure 3. Immature female LDLR–/– mice were treated daily with vehicle control or VCD (160 mg/kg per day IP for 15 d) or were ovariectomized and received 17?-estradiol supplementation as indicated. Plasma samples were collected as described in Methods and total cholesterol levels determined (*P<0.05).
TABLE 1. Lipid Profiles
Endocrine Effects
On day 120 after the onset of VCD treatment, a reduction (P<0.05) in ovarian weight was observed in VCD and VCD+E2-treated mice compared with vehicle control (Figure 4A). The weight of ovaries from VCD-treated mice was 34% of control, whereas ovaries from VCD+E2 were 23% of control (control 0.000235±0.0001; VCD 0.00080±0.0001; VCD+E2 0.000053±0.00001 gm/gm body weight; Figure 4A). Furthermore, on histological examination, VCD-treated ovaries had no ovarian follicles compared with ovaries from vehicle control cycling animals, which contained follicles at all stages of development and corpora lutea (data not shown).
Figure 4. Immature female LDLR–/– mice were treated daily with vehicle control or VCD (160 mg/kg per day IP for 15 d) or were ovariectomized and received 17?-estradiol supplementation as indicated. Tissues were collected, weighed, and wet weights recorded on day 120 after onset of dosing. A, Ovarian weights (n=11 to 13). B, Uterine weights (n=5 to 13; *P<0.05).
Uterine weight in OVX mice was reduced (P<0.05) compared with vehicle control (Figure 4B). Uterine weight in VCD-treated mice was not different from vehicle control mice (Figure 4B). VCD+E2 and OVX+E2 mice had increased (P<0.05) uterine weights compared with VCD and OVX uterine weights.
On day 53 after the onset of VCD dosing, plasma FSH levels were greater (P<0.05) in OVX and VCD-treated mice relative to vehicle-treated cycling controls (Table 2). Plasma FSH levels in OVX mice were not significantly different from VCD-treated mice. Estradiol supplementation reduced (P<0.05) plasma FSH in VCD+E2 and OVX+E2 mice, but plasma FSH levels were not as low as the level observed in the cycling vehicle control mice (Table 2).
TABLE 2. Plasma Levels of Steroid Hormones and Gonadotropins
Plasma levels of 17?-estradiol, measured in a cumulative plasma pool as described in Methods, were nondetectable in vehicle control, OVX, and VCD-treated animals. Whereas plasma levels of estradiol in the OVX+E2 mice were increased above physiological levels, the estradiol level in VCD+E2-treated mice was within the physiological range during the estrus cycle. Plasma levels of E2 in various stages of cycling LDLR–/– mice in a baseline cohort were 82.6±48.0 pg/mL (n=6). The plasma estradiol levels were lower (P<0.05) in VCD+E2 mice compared with OVX+E2 mice.
Plasma levels of androstenedione were similar in vehicle and VCD-treated mice measured in a cumulative plasma pool, were reduced (P<0.05) in OVX mice compared with vehicle control mice, and reduced (P<0.05) in VCD+E2 mice compared with mice treated with only VCD (Table 2).
Discussion
The aim of this study was to compare development of atherosclerotic lesions in VCD-treated and OVX±E2 mice. The rationale was based on comparing lifetime hormone profiles and temporal patterns of ovarian failure between individuals who undergo surgical (OVX) versus natural (VCD-treated) menopause. Here, in LDLR–/– mice, protection against lesion development was provided by E2 supplementation in animals given hormone before complete ovarian failure induced by VCD. In fact, E2 supplementation before or during onset of follicle depletion appeared to be more beneficial because innominate lesion area was significantly less in VCD+E2 compared with OVX+E2. This result is consistent with observations in monkeys in which E2 supplementation at the time of OVX is protective against CVD development, whereas it is detrimental if given after cardiovascular lesion progression is under way.17 Our findings suggest the timing of estrogen replacement may be critical to the subsequent influence on atherosclerosis progression, as suggested recently by Dubey et al,18 and may reflect tissues that continue to respond because they have not been estrogen deficient for long periods of time.19
The data presented here are consistent with previous findings of E2-mediated lesion area reduction independent of alterations in lipid levels.16 Here, aortae lesion development was significantly less in E2-supplemented OVX and VCD-treated groups; however, in the innominate artery, there was a significant difference in atherosclerotic lesion area. Recent studies suggest the innominate artery may be more relevant for investigating the rate of lesion progression and features that contribute to clinically significant disease.20–22 Striking was the effect of E2 supplementation on reducing innominate lesion in the VCD-treated mice compared with OVX mice. Perhaps gradual loss of estrogen, which could include spiking of estrogen to levels higher than in normal estrus cycling in the VCD-treated mouse, added to the atheroprotective effect of the E2 supplementation when compared with the abrupt changes observed in the OVX mice. Circulating androgen in the VCD-treated animal could be aromatized in the vessel wall to exert a localized protective effect of estrogen.23 Nor can we rule out the possibility of a direct effect of circulating androgen on reducing the carotid intimal-medial thickness, which has been reported for women.24
VCD-induced premature ovarian failure in mice leads to the development of a follicle-depleted, ovary-intact animal that resembles the dynamic endocrine state in perimenopausal and postmenopausal women.14 Perimenopause is defined as the period of 1 to 10 years before the final menstrual period, when symptoms of menopause begin and many disease risks increase.25,26 The VCD-induced follicle-depletion protocol was used in the LDLR–/– atherosclerosis model to mimic conditions observed in a healthy, perimenopausal population. This made it possible to determine the effects of estrogen supplementation on the development of atherosclerosis in mice undergoing gradual loss of ovarian-derived 17?-estradiol (VCD-treated) versus ovariectomized mice, with an abrupt withdrawal of ovarian function coincident with atherosclerotic lesion development.
There was no evidence of any generalized toxic effect of VCD treatment on the LDLR–/– C57BL/6J mice. This finding is consistent with previous findings in the hybrid B6C3F114 and inbred C57BL/6J and C3H strains (Kitten AM, unpublished data, 2003). Nor were there significant differences in plasma total cholesterol, HDL cholesterol, or triglycerides with VCD treatment, another indication that liver function was not compromised by VCD treatment. At the end of the experiment on day 120, there was no change in plasma total cholesterol of VCD-treated mice relative to OVX mice, indicating that differences in innominate lesion area were not attributable to plasma lipids.
VCD treatment caused a complete loss of all stages of ovarian follicles by day 120 after the onset of treatment, confirming premature ovarian failure had been induced. VCD-induced follicle loss and ovarian weight were not affected by E2 supplementation. Conversely, uterine weight, a physiological measure of uterotrophic effects of E2, was significantly greater in response to estrogen supplementation in OVX and VCD-treated animals. Uterine weight in VCD-treated mice was equivalent to that in intact cycling mice and significantly greater than OVX mice, which was apparently because of androgen promotion of uterine weight increase.27,28 The lack of differences in uterine weights between VCD-treated animals and vehicle cycling controls may be attributable to a strain difference or variation in collection methods from our previous report.14
FSH levels increased in LDLR–/– transgenic mice similar to VCD-induced follicle loss in other strains of mice.14,15 E2 supplementation of OVX and VCD mice caused FSH to drop, but not to levels observed in cycling mice. This supports that exogenous E2 was only partially effective at invoking negative feedback on pituitary FSH secretion. FSH release by the pituitary is also inhibited by ovarian-derived inhibin, a member of the transforming growth factor-? family of peptides.29 Inhibin is produced by ovarian granulosa cells,30 and, without granulosa cells, whether in OVX mice or follicle-depleted ovaries of VCD mice, the absence of inhibin was reflected in the FSH levels being reduced by half with E2 treatment but still significantly greater than in the cycling, intact mice.
The cumulative steroid pool collected throughout the study was designed to mimic the cumulative hormone exposure a woman would experience during the transition from reproductive years into the perimenopause and postmenopause periods.25,26 It has been reported that the rodent adrenal does not produce androgen.29,31 However, our results in the OVX mice, as well as those of Villablanca in castrated male mice,32 suggest an extragonadal source of androgen in mice. In our previous study, plasma androstenedione measured at a single time point on day 120 after initiation of VCD treatment was one-third the concentration in a cycling control animal.14 However, in the study reported here, androstenedione levels were measured in a pooled cumulative sample, which may explain why the level was not different between vehicle control and VCD-treated mice. E2 supplementation significantly reduced androstenedione in the VCD+E2 follicle-depleted mice compared with an increase in androstenedione seen in OVX+E2 mice. Interstitial cells of the follicle-depleted ovarian tissue produce androstenedione in culture, which is blocked by E2.33 We conclude that E2 supplementation in VCD-treated mice reduced ovarian interstitial cell production of androstenedione.
An additional endocrine difference was decreased, circulating estradiol in VCD+E2-treated mice compared with OVX+E2 mice. Estrogen is inactivated and removed by sulfoconjugation catalyzed by estrogen sulfotransferase, the expression of which in the ovary is stimulated by LH, the gonadotropin that stimulates interstitial cells.34,35 The follicle-depleted ovary enriched in interstitial cells may metabolize estradiol via estrogen sulfotransferase, resulting in less estradiol in the VCD+E2 mice compared with the OVX+E2 mice.
Together, our findings suggest that studies using the VCD-treated mouse as a model for CVD and menopause may provide an endocrine milieu more like that in perimenopausal and postmenopausal women. The VCD model would be better to assess the potential impact of therapeutic interventions in primary or secondary CVD prevention, for instance, to investigate the role of cyclooxygenase 2 inhibitors,36 a subject of recent investigations. Furthermore, using a more endocrine-relevant model will enable the performance of animal studies designed to reconcile the differences reported in clinical investigations regarding the risks or benefits of hormone supplementation.2–6,18 Another uncertainty that could be addressed using the VCD model of ovarian failure is the role of androgens in CVD. It has been assumed for decades that androgen is atherogenic and estrogen atheroprotective but, careful re-examination suggests that androgen and estrogen are atheroprotective.37–39 The VCD mouse model with sustained androgen and progesterone production from the follicle-depleted ovary provides a physiological setting in which to identify the roles of these steroids in atherosclerotic lesion development. Finally, the potential utility of this model for screening therapeutic approaches will likely facilitate drug discovery in women’s health and, more importantly, provide an early warning for contraindications of drugs prescribed for women in perimenopausal years.
Acknowledgments
This work was supported by the American Heart Association fellowship, ATVB junior investigator for women 2003 (L.P.M.); and grants ES08979, ES09246, and AG21948 (P.B.H.), center grant ES-06694 (P.B.H.), and HL55517 (C.L.B.). The authors thank Patrick J. Devine for his continued support and comments, and Patricia J. Christian, David Ruecker, Meriah Earle, Eric Cheung, Samuel Marion, and Virginia S. Patch for technical assistance. The authors wish to extend a special acknowledgement to Susan D. Airhart and Allison Kitten from the Jackson Laboratory for stimulating discussions and continued support.
References
American Heart Association web site. http://www.americanheart.org/downloadable/heart/1104938287787FS10WM05rev.pdf.
Gordon T, Kannel WB, Hjortland MC, McNamara PM. Menopause and coronary heart disease. The Framingham Study. Ann Intern Med. 1978; 89: 157–161.
Kritz-Silverstein D, Barrett-Connor E, Wingard DL. Hysterectomy, oophorectomy, and heart disease risk factors in older women. Am J Public Health. 1997; 87: 676–680.
Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J; Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. J Am Med Assoc. 2002; 288: 321–333.
Grodstein F, Manson JE, Colditz GA, Willett WC, Speizer FE, Stampfer MJ. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med. 2000; 133: 933–941.
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vittinghoff E. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. J Am Med Assoc. 1998; 280: 605–613.
Haas JS, Kaplan CP, Gerstenberger EP, Kerlikowske K. Changes in the use of postmenopausal hormone therapy after the publication of clinical trial results. Ann Intern Med. 2004; 140: 184–188.
Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, Bonds D, Brunner R, Brzyski R, Caan B, Chlebowski R, Curb D, Gass M, Hays J, Heiss G, Hendrix S, Howard BV, Hsia J, Hubbell A, Jackson R, Johnson KC, Judd H, Kotchen JM, Kuller L, LaCroix AZ, Lane D, Langer RD, Lasser N, Lewis CE, Manson J, Margolis K, Ockene J, O’Sullivan MJ, Phillips L, Prentice RL, Ritenbaugh C, Robbins J, Rossouw JE, Sarto G, Stefanick ML, Van Horn L, Wactawski-Wende J, Wallace R, Wassertheil-Smoller S; Women’s Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. J Am Med Assoc. 2004; 291: 1701–1712.
Hodgin JB, Maeda N. Minireview: estrogen and mouse models of atherosclerosis. Endocrinology. 2002; 143: 4495–4501.
Holm P. Effect of estrogen on development of atherosclerosis. A review of experimental animal studies. Dan Med Bull. 2001; 48: 146–160.
Bellino FL. Nonprimate animal models of menopause: workshop report. Menopause. 2000; 7: 14–24.
Mansfield PK, Carey M, Anderson A, Barsom SH, Koch PB. Staging the menopausal transition: data from the TREMIN Research Program on Women’s Health. Womens Health Issues. 2004; 14: 220–226.
North American Menopause Society (NAMS). Menopause Practice: A Clinician’s Guide. NAMS 2004 PDF on web site http://www.menopause.org.
Mayer L, Devine PJ, Dyer CA, Hoyer PB. The follicle-deplete ovary produces androgen. Biol Reprod. 2004; 71: 130–138.
Hooser SB, Douds DP, DeMerrel DG, Hoyer PB, Sipes IG. Long-term ovarian and gonadotropin changes in mice exposed to 4-vinylcyclohexene. Reprod Toxicol. 1994; 8: 315–323.
Marsh MM, Walker VR, Curtiss LK, Banka CL. Protection against atherosclerosis by estrogen is independent of plasma cholesterol levels in LDL receptor-deficient mice. J Lipid Res. 1999; 40: 893–900.
Kaplan JR, Manuck SB, Anthony MS, Clarkson TB. Premenopausal social status and hormone exposure predict postmenopausal atherosclerosis in female monkeys. Obstet Gynecol. 2002; 99: 381–388.
Dubey RK, Imthurn B, Zacharia LC, Jackson EK. Hormone replacement therapy and cardiovascular disease:what went wrong and where do we go from here? Hypertension. 2004; 44: 789–795.
Gambrell RD Jr. The women’s health initiative reports in perspective:facts or fallacies? Climacteric. 2004; 7: 221–224.
Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, Schwartz SM. Advanced atherosclerotic lesions in the innominate artery of the Apo E knockout mouse. Arterioscler Thromb Vasc Biol. 2000; 20: 2587–2592.
Reardon CA, Getz GS. Mouse models of atherosclerosis. Curr Opin Lipidol. 2001; 12: 167–173.
Hockings PD, Roberts T, Galloway GJ, Reid DG, Harris DA, Vidgeon-Hart M, Groot PH, Suckling KE, Benson GM Repeated three-dimensional magnetic resonance imaging of atherosclerosis development in innominate arteries of low-density lipoprotein receptor-knockout mice. Circulation. 2002; 106: 1716–1721.
Nathan L, Shi W, Dinh H, Mukherjee TK, Wang X, Lusis AJ, Chaudhuri G. Testosterone inhibits early atherogenesis by conversion to estradiol: critical role of aromatase. Proc Natl Acad Sci U S A. 2001; 98: 3589–3593.
Bernini GP, Sgro M, Moretti A, Argenio GF, Barlascini CO, Cristofani R, Salventti A. Endogenous androgens and carotid intimal-medial thickness in women. J Clin Endocrinol Metab. 1999; 84: 2008–2012.
Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab. 1996; 81: 1495–1501.
Burger HG, Dudley EC, Robertson DM, Dennerstein L. Hormonal changes in the menopause transition. Recent Prog Horm Res. 2002; 57: 257–275.
Schmidt WN, Katzenellenbogen BS. Androgen-uterine interactions: an assessment of androgen interaction with the testosterone- and estrogen-receptor systems and stimulation of uterine growth and progesterone-receptor synthesis. Mol Cell Endocrinol. 1979; 15: 91–108.
Yamasaki K, Takeyoshi M, Sawaki M, Imatanaka N, Shinoda K, Takatsuki M. Immature rat uterotrophic assay of 18 chemicals and Hershberger assay of 30 chemicals. Toxicology. 2003; 183: 93–115.
van Weerden WM, Bierings HG, van Steenbrugge GJ, de Jong FH, Schroder FH. Adrenal glands of mouse and rat do not synthesize androgens. Life Sci. 1992; 50: 857–861.
Hurwitz JM, Santoro N. Inhibins, activins, and follistatin in the aging female and male. Semin Reprod Med. 2004; 22: 209–217.
Dufau ML, Miyagawa Y, Takada S, Khanum A, Miyagawa H, Buczko E. Regulation of androgen synthesis: the late steroidogenic pathway. Steroids. 1997; 62: 128–132.
Villablanca A, Lubahn D, Shelby L, Lloyd K, Barthold S. Susceptibility to early atherosclerosis in male mice is mediated by estrogen receptor alpha. Arterioscler Thromb Vasc Biol. 2004; 1055–1061.
Magoffin DA, Erickson GF. Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium. Mol Cell Endocrinol. 1982; 28: 81–89.
Tong MH, Christenson LK, Song WC. Aberrant cholesterol transport and impaired steroidogenesis in Leydig cells lacking estrogen sulfotransferase. Endocrinology. 2004; 145: 2487–2497.
Jo M, Gieske MC, Payne CE, Wheeler-Price SE, Gieske JB, Ignatius IV, Curry TE Jr, Ko C. Development and application of a rat ovarian gene expression database. Endocrinology. 2004; 145: 5384–5396.
Egan KM, Lawson JA, Fries S, Koller B, Rader DJ, Smyth EM, Fitzgerald GA. COX-2-derived prostacyclin confers atheroprotection on female mice. Science. 2004; 306: 1954–1957.
Liu PY, Death AK, Handelsman DJ. Androgens and cardiovascular disease. Endocr Rev. 2003; 24: 313–340.
Wu FC, von Eckardstein A. Androgens and coronary artery disease. Endocr Rev. 2003; 24: 183–217.
Phillips GB. Is atherosclerotic cardiovascular disease an endocrinological disorder? The estrogen-androgen paradox. J Clin Endocrinol Metab. 2005; 90: 2708–2711.(Loretta P. Mayer; Cheryl )
Reprint requests to Loretta P. Mayer, PhD, Northern Arizona University, Department of Biological Sciences, Box 5640, Flagstaff, AZ 86011-5640. E-mail Loretta.Mayer@nau.edu
Abstract
Objective— Since the unexpected results from the Women’s Health Initiative, the possible protective role of estrogen in preventing heart disease in perimenopausal and postmenopausal women is uncertain. This study examined atherosclerotic lesion development in ovariectomized versus follicle-depleted ovary-intact cholesterol-fed female low-density lipoprotein (LDL) receptor–deficient mice.
Methods and Results— We studied lesion development in LDL receptor–deficient mice that were ovariectomized or follicle depleted with 4-vinylcyclohexene diepoxide (VCD) to induce ovarian failure, then treated ± exogenous 17?-estradiol via pellet implant. At 120 days after start of cholesterol feeding, the extent of lesion in aorta and innominate artery was determined. Lesion area in both locations was similar in vehicle control, VCD-treated, and ovariectomized mice. Replacement with 17?-estradiol caused lesion reduction (P<0.05) in both arterial locations, but it was most efficacious in suppressing innominate lesion area in VCD-treated mice (12.9±5.2%) compared with ovariectomized mice (40.0±6.04%).
Conclusions— Endocrine status associated with the follicle-depleted ovary influences exogenous estradiol effects during the development of atherosclerotic lesions and, in particular, inhibits lesion progression in the innominate artery.
The goal was to determine whether estrogen treatment altered atherosclerosis development in follicle-deplete ovary-intact mice compared with ovariectomized mice. Ovarian contribution of androgen appeared to augment the protective effects of estrogen, suggesting this new model may provide an improved experimental system to analyze atherosclerosis drugs for perimenopausal and postmenopausal women.
Key Words: atherosclerosis ? estrogen supplementation ? perimenopause model
Introduction
Despite women being protected from cardiovascular disease (CVD) until well after menopause, CVD kills more women than men every year in the United States.1 It was presumed for decades that the atheroprotection enjoyed by women was primarily attributable to ovarian estradiol production. Once ovarian production of estradiol stopped because of surgical or spontaneous menopause, protection was lost, and CVD progressed to clinical complications.2,3 On the basis of this presumption, estrogen replacement therapy was initiated during the 1960s and was predicted to be atheroprotective until July 2002, with the announcement of the unexpected findings of increased CVD in the combined hormone therapy arm of the randomized, controlled Women’s Health Initiative (WHI) trial.4
See page 1765
The WHI results contradicted previous observational studies, such as the Nurses’ Health Study, that reported significant protection from CVD with hormone replacement5 but was consistent with the Heart and Estrogen/progestin Replacement Study (HERS) that reported estrogen replacement did not provide protection to women with existing CVD.6 Nonetheless, the WHI results were surprising in providing evidence that combined hormone replacement may be detrimental in healthy women and had a dramatic impact causing many women to stop hormone replacement therapy.7 In February 2004, the estrogen-only arm of WHI was stopped. Interestingly, unlike the findings in women receiving combined estrogen progestin therapy, there was no increase in cardiovascular events in women treated with conjugated equine estrogens alone.8
The putative atheroprotection of estrogen reported in observational human studies is supported by numerous studies demonstrating protection in a variety of animal models of CVD.9,10 Because only human females have extended postmenopausal survival (as much as one third of the life span is postreproductive), there are no appropriate age-dependent animal models of menopause, and studies rely on surgical removal of the ovaries to mimic the endocrine status of postmenopausal women.11 Surgical menopause (total hysterectomy) obviates the perimenopausal transition that can take 10 years and involves multiple ovarian hormonal changes with time.12 Surgical menopause occurs in <13% of postmenopausal women.13 Therefore, the majority of women experience menopause through a natural transition resulting from ovarian aging and follicle depletion and have residual ovarian tissue devoid of estrogen-producing follicles that can produce progesterone and androgen.3
Recently, we reported a method to chemically induce follicle depletion in mice by accelerating the natural process of atresia.14 The industrial chemical 4-vinylcyclohexene diepoxide (VCD) selectively targets primordial and primary ovarian follicles.14 Because larger-growing follicles, the source of ovarian estrogen, are not targeted by VCD, their depletion is only manifest after all nonrenewing primordial follicles are depleted and can no longer serve as a pool for recruitment to larger follicles. Once ovarian estrogen and inhibin production ceases as a result of follicle depletion, the negative feedback to the pituitary is gone, and plasma follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels rise.14,15 In response to elevated LH, cells in the residual ovarian tissue, likely interstitial in origin, continue to produce steroids: progesterone and androgen, primarily.14 Because ovarian follicles are lost gradually and the animal retains residual steroidogenic ovarian tissue, the VCD-treated mouse has an endocrine profile more closely parallel to that in human perimenopause and postmenopause than that in ovariectomized mouse.14
We reported previously that 17?-estradiol replacement resulted in a significant reduction of atherosclerotic lesions in ovariectomized low-density lipoprotein receptor–deficient (LDLR–/–) mice fed a high-fat, high-cholesterol diet, and the effect was unrelated to plasma cholesterol levels.16 The present study was designed to compare the effect of 17?-estradiol treatment on lesion formation in VCD-induced follicle-depleted, ovary-intact mice and ovariectomized mice to determine whether the differences in endocrine milieu altered the extent of lesions formed in the aortic and innominate arteries.
Methods
Animals
C57BL/6J LDLR–/– mice were purchased from The Jackson Laboratory (Bar Harbor, Maine) and bred in the Banka laboratory. All studies were approved by the La Jolla Institute for Molecular Medicine institutional animal care and use committee and comply with National Institutes of Health guidelines. Mice (75 total) were weaned at 28 days (d), initially fed ad libitum a standard mouse chow diet (Diet No. 5015; Harlan Teklad), housed 3 per cage in autoclaved filter-top cages with autoclaved water, and kept on a 12-hour light/dark cycle in a conventional clean room facility. The treatment protocol for VCD induction of follicular depletion has a patent pending (The University of Arizona). VCD-treated mice are commercially available through The Jackson Laboratory (Bar Harbor, Me). As illustrated in Figure 1, sexually immature mice (28 to 35 d old) were randomly selected for each treatment group, weighed, and injected intraperitoneal daily with VCD (Sigma-Aldrich) in sesame oil at 160 mg/kg per day or sesame oil vehicle control for 15 d (n=6 to 13 per treatment group). At sexual maturity (58 d old), bilateral ovariectomy (OVX) was performed on 2 groups of sesame oil–treated mice (n=6 to 9 per group). Mice were anesthetized for surgery using inhalational methoxyflurane (Metofane; Malinckrodt Veterinary) and intraperitoneal ketamine (Ketostat; 100 mg/kg; Ft Dodge Laboratories Inc).
Figure 1. The diagram represents the study as designed to compare endocrine and cardiovascular attributes in normal cycling, VCD-treated, and ovariectomized LDLR–/– mice on a high-fat diet. Treatments and sample collections were performed as indicated.
Female OVX and VCD-treated mice were implanted on day 32 after the onset of VCD treatment with subcutaneous slow-release hormone pellets (Innovative Research of America) designed to release 4 μg per day for 90 d of 17?-estradiol (E2; n=9 to 13). Three days after OVX surgery and day 33 after the onset of VCD treatment, all mice were started on an atherogenic high-cholesterol diet containing 1.25% cholesterol, 6% fat, minimum essential dietary requirements of vitamin E, and no cholate for 12 weeks; day 120 (Diet No. TD96335; Harlan Teklad).
Tissues
Heparinized venous blood was collected from the retro-orbital sinus under anesthesia after an 8-hour fast unless otherwise specified. Blood samples were collected from animals on a chow diet to correspond to animals on high-fat diet collected at 25, 53, 81, 109, and 120 d after the onset of VCD treatment (Figure 1). The only nonfasting samples were collected at killing (90 d after surgery and 120 d after the onset of VCD treatment). Plasma was obtained by centrifugation and stored (–20°C) before analysis. Mice were weighed while anesthetized. A cumulative plasma pool was generated for steroid hormone measurements by pooling 125 μL of plasma from each blood sample collection time point on day 25, 53, 81, 109, and 120 after the onset of VCD treatment.
On day 120 after VCD treatment, animals were weighed, bled, and perfused under anesthesia. An incision was made in the inferior vena cava, and a cannula was inserted into the left ventricle for perfusion with PBS followed by formal sucrose (4% paraformaldehyde and 5% sucrose in PBS; Sigma-Aldrich). The adrenals, spleen, uteri, livers, kidneys, and ovaries were collected.
All organs were trimmed free of fat and wet weights recorded. All organ weights were normalized to body weight to correct for differences in total body weight. Uteri were excised from the oviduct at the uterotubal junction bilaterally and transected at the junction of the uterine body and cervix. After removal of luminal fluid, the entire uterus was weighed on a microbalance. One ovary from each treatment group was prepared for histological evaluation (n=6 animals per group). The ovary was placed in 4% paraformaldehyde fixative, transferred to 70% ethanol, paraffin embedded and serially sectioned (4 to 5 μm), mounted, and stained with hematoxylin and eosin. Follicles were counted in every twentieth section and classified as described previously.14
Innominate arteries were isolated, cleaned, and photographed in situ under the dissecting microscope using an Olympus DP11 digital camera. Photographic images were subsequently analyzed morphometrically by drawing a line across the origin from the aorta and a parallel line intersecting the bifurcation to the subclavian and common carotid arteries (Figure 2, red lines). This defined the total area of the innominate artery. Lesion area was outlined and the percentage of the total area covered by lesion calculated using NIH Image software. Using a dissecting microscope, the entire mouse aorta was dissected, removed, and prepared as described previously.16 The Sudan IV–stained aortae were photographed at a fixed magnification using an Olympus DP11 and imported into a computer for morphometric analysis (NIH Image software) as described previously.16
Figure 2. Immature female LDLR–/– mice were treated daily with vehicle control, VCD (160 mg/kg per day IP for 15 d; n=6 to 12), or were ovariectomized and received 17?-estradiol supplementation as indicated. Innominate arteries (A) and aortae (B) were collected and prepared on day 120 after onset of dosing (*P<0.05 different from control; letters different within group).
Biochemical Analyses
Hepatocellular histopathology and enzymatic activity of circulating aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were determined by the diagnostic laboratory at the Arizona Health Sciences Center. Total cholesterol and triglycerides were quantified using Sigma reagents 401-25P and 344-20, respectively. High-density lipoprotein (HDL) cholesterol was measured using the Sigma 401-25P reagent after precipitation of the apolipoprotein B–containing lipoproteins (reagent 543004; Boehringer Mannheim).
Plasma levels of 17?-estradiol and androstenedione were determined by Coat-A-Count radioimmunossay (RIA; Diagnostic Products Inc.). Sensitivity and interassay and intra-assay coefficients of variation were 17?-estradiol, 1.4 pg/mL, 4.6% and 14%; androstenedione, 0.04 ng/mL, 8.4% and 5.7%, respectively. Plasma FSH was measured by RIA according to instructions with the murine kit from the National Hormone and Pituitary Distribution Program. All samples were assayed in duplicate. Sensitivity and interassay and intra-assay coefficients of variation were 200 pg/mL, 2.7%, and 6.7%. The results of all RIAs were calculated by 4-parameter logistic analysis using the software Assay Zap (BioSoft).
Data Analysis
Body weights, tissue weights, and plasma hormone concentrations were determined from individual animals and averaged for each treatment, and the means (±SEM) in control versus treated animals were analyzed for significant differences by 1-way ANOVA with significance set at P<0.05. Post hoc tests were used where appropriate, with Dunnett’s used to determine differences from control and Student t test to compare each pair. Tests for homogeneity of variance (Bartlett) and normality (Shapiro–Wilk) were routinely performed to ensure that the assumptions of the ANOVA were met. Lipid measurements were analyzed using a Friedman repeated-measures ANOVA on ranks. The Student-Newman–Keuls method was used for pairwise multiple comparisons of groups as a post hoc test.
Results
Innominate Atherosclerotic Lesion Area
The extent of innominate lesion area was similar and not significantly different among control (47.3±4.8%), VCD (45.04±5.7%), and OVX (57.0±7.3%) treatment groups ranging from 40% to 60% (Figure 2A). However, E2 supplementation in VCD-treated mice resulted in significant differences in lesion area compared with that in OVX mice. The VCD+E2 (12.9±5.2%) treatment group had less (P<0.05) lesion area than the OVX+E2 (40.0±6.0%) treatment group (n=6 to 12).
Aortic Atherosclerotic Lesion Area
The extent of aortic lesion area was similar and not significantly different among the control (6.4±0.7%), VCD (6.8±0.7%), and OVX (8.3±1.5%) groups, ranging from 5% to 8% (Figure 2B). In VCD+E2 (1.3±0.9%) and OVX+E2 (1.7±0.7%), there was a reduction (P<0.05) in aortic lesion area; however, the extent of E2 effects in these 2 treatment groups was not different (n=6 to 12).
Systemic and Liver Function
Body weights at day 120 did not differ significantly among the 5 groups: vehicle, VCD, VCD+17?-estradiol (E2), OVX, OVX+E2 (data not shown). There was no effect of either VCD±E2 or OVX±E2 on adrenal, kidney, or spleen weights normalized to body weight (data not shown). Levels of liver enzymes AST and ALT were within the normal range for VCD-treated and control animals (AST; control, 157.85±56.55 U/L; VCD, 57.30±2.30 U/L: normal range 54 to 298 U/L: ALT; control, 34.53±10.31 U/L; VCD, 35.84±12.40 U/L: normal range 17 to 77 U/L). ALT and AST circulating levels were not analyzed in OVX±E2 mice. There was no histological evidence of hepatocellular pathology in livers from VCD±E2 or OVX±E2 mice (data not shown).
Lipids
Over the time course of the experiment, once fat feeding had commenced, plasma total cholesterol increased 3- to 4-fold from day 25 to day 53 in all experimental groups: vehicle, VCD, VCD+E2, OVX, and OVX+E2 (Figure 3). On day 53, total cholesterol was significantly less in VCD+E2, OVX, and OVX+E2 compared with vehicle control mice. On day 81, plasma total cholesterol was elevated in VCD and OVX mice compared with vehicle control. At day 120, OVX mice had significantly elevated cholesterol compared with the other treatment groups, and there was a significant reduction in plasma cholesterol in OVX+E2 mice. On day 120, neither HDL nor triglycerides differed among any of the groups, suggesting that the significant decrease observed in plasma total cholesterol in OVX+E2 was attributable to reduction in LDL cholesterol because there were no significant reductions in plasma triglycerides, indicating that the affected fraction was LDL and not very–low-density lipoprotein (Table 1).
Figure 3. Immature female LDLR–/– mice were treated daily with vehicle control or VCD (160 mg/kg per day IP for 15 d) or were ovariectomized and received 17?-estradiol supplementation as indicated. Plasma samples were collected as described in Methods and total cholesterol levels determined (*P<0.05).
TABLE 1. Lipid Profiles
Endocrine Effects
On day 120 after the onset of VCD treatment, a reduction (P<0.05) in ovarian weight was observed in VCD and VCD+E2-treated mice compared with vehicle control (Figure 4A). The weight of ovaries from VCD-treated mice was 34% of control, whereas ovaries from VCD+E2 were 23% of control (control 0.000235±0.0001; VCD 0.00080±0.0001; VCD+E2 0.000053±0.00001 gm/gm body weight; Figure 4A). Furthermore, on histological examination, VCD-treated ovaries had no ovarian follicles compared with ovaries from vehicle control cycling animals, which contained follicles at all stages of development and corpora lutea (data not shown).
Figure 4. Immature female LDLR–/– mice were treated daily with vehicle control or VCD (160 mg/kg per day IP for 15 d) or were ovariectomized and received 17?-estradiol supplementation as indicated. Tissues were collected, weighed, and wet weights recorded on day 120 after onset of dosing. A, Ovarian weights (n=11 to 13). B, Uterine weights (n=5 to 13; *P<0.05).
Uterine weight in OVX mice was reduced (P<0.05) compared with vehicle control (Figure 4B). Uterine weight in VCD-treated mice was not different from vehicle control mice (Figure 4B). VCD+E2 and OVX+E2 mice had increased (P<0.05) uterine weights compared with VCD and OVX uterine weights.
On day 53 after the onset of VCD dosing, plasma FSH levels were greater (P<0.05) in OVX and VCD-treated mice relative to vehicle-treated cycling controls (Table 2). Plasma FSH levels in OVX mice were not significantly different from VCD-treated mice. Estradiol supplementation reduced (P<0.05) plasma FSH in VCD+E2 and OVX+E2 mice, but plasma FSH levels were not as low as the level observed in the cycling vehicle control mice (Table 2).
TABLE 2. Plasma Levels of Steroid Hormones and Gonadotropins
Plasma levels of 17?-estradiol, measured in a cumulative plasma pool as described in Methods, were nondetectable in vehicle control, OVX, and VCD-treated animals. Whereas plasma levels of estradiol in the OVX+E2 mice were increased above physiological levels, the estradiol level in VCD+E2-treated mice was within the physiological range during the estrus cycle. Plasma levels of E2 in various stages of cycling LDLR–/– mice in a baseline cohort were 82.6±48.0 pg/mL (n=6). The plasma estradiol levels were lower (P<0.05) in VCD+E2 mice compared with OVX+E2 mice.
Plasma levels of androstenedione were similar in vehicle and VCD-treated mice measured in a cumulative plasma pool, were reduced (P<0.05) in OVX mice compared with vehicle control mice, and reduced (P<0.05) in VCD+E2 mice compared with mice treated with only VCD (Table 2).
Discussion
The aim of this study was to compare development of atherosclerotic lesions in VCD-treated and OVX±E2 mice. The rationale was based on comparing lifetime hormone profiles and temporal patterns of ovarian failure between individuals who undergo surgical (OVX) versus natural (VCD-treated) menopause. Here, in LDLR–/– mice, protection against lesion development was provided by E2 supplementation in animals given hormone before complete ovarian failure induced by VCD. In fact, E2 supplementation before or during onset of follicle depletion appeared to be more beneficial because innominate lesion area was significantly less in VCD+E2 compared with OVX+E2. This result is consistent with observations in monkeys in which E2 supplementation at the time of OVX is protective against CVD development, whereas it is detrimental if given after cardiovascular lesion progression is under way.17 Our findings suggest the timing of estrogen replacement may be critical to the subsequent influence on atherosclerosis progression, as suggested recently by Dubey et al,18 and may reflect tissues that continue to respond because they have not been estrogen deficient for long periods of time.19
The data presented here are consistent with previous findings of E2-mediated lesion area reduction independent of alterations in lipid levels.16 Here, aortae lesion development was significantly less in E2-supplemented OVX and VCD-treated groups; however, in the innominate artery, there was a significant difference in atherosclerotic lesion area. Recent studies suggest the innominate artery may be more relevant for investigating the rate of lesion progression and features that contribute to clinically significant disease.20–22 Striking was the effect of E2 supplementation on reducing innominate lesion in the VCD-treated mice compared with OVX mice. Perhaps gradual loss of estrogen, which could include spiking of estrogen to levels higher than in normal estrus cycling in the VCD-treated mouse, added to the atheroprotective effect of the E2 supplementation when compared with the abrupt changes observed in the OVX mice. Circulating androgen in the VCD-treated animal could be aromatized in the vessel wall to exert a localized protective effect of estrogen.23 Nor can we rule out the possibility of a direct effect of circulating androgen on reducing the carotid intimal-medial thickness, which has been reported for women.24
VCD-induced premature ovarian failure in mice leads to the development of a follicle-depleted, ovary-intact animal that resembles the dynamic endocrine state in perimenopausal and postmenopausal women.14 Perimenopause is defined as the period of 1 to 10 years before the final menstrual period, when symptoms of menopause begin and many disease risks increase.25,26 The VCD-induced follicle-depletion protocol was used in the LDLR–/– atherosclerosis model to mimic conditions observed in a healthy, perimenopausal population. This made it possible to determine the effects of estrogen supplementation on the development of atherosclerosis in mice undergoing gradual loss of ovarian-derived 17?-estradiol (VCD-treated) versus ovariectomized mice, with an abrupt withdrawal of ovarian function coincident with atherosclerotic lesion development.
There was no evidence of any generalized toxic effect of VCD treatment on the LDLR–/– C57BL/6J mice. This finding is consistent with previous findings in the hybrid B6C3F114 and inbred C57BL/6J and C3H strains (Kitten AM, unpublished data, 2003). Nor were there significant differences in plasma total cholesterol, HDL cholesterol, or triglycerides with VCD treatment, another indication that liver function was not compromised by VCD treatment. At the end of the experiment on day 120, there was no change in plasma total cholesterol of VCD-treated mice relative to OVX mice, indicating that differences in innominate lesion area were not attributable to plasma lipids.
VCD treatment caused a complete loss of all stages of ovarian follicles by day 120 after the onset of treatment, confirming premature ovarian failure had been induced. VCD-induced follicle loss and ovarian weight were not affected by E2 supplementation. Conversely, uterine weight, a physiological measure of uterotrophic effects of E2, was significantly greater in response to estrogen supplementation in OVX and VCD-treated animals. Uterine weight in VCD-treated mice was equivalent to that in intact cycling mice and significantly greater than OVX mice, which was apparently because of androgen promotion of uterine weight increase.27,28 The lack of differences in uterine weights between VCD-treated animals and vehicle cycling controls may be attributable to a strain difference or variation in collection methods from our previous report.14
FSH levels increased in LDLR–/– transgenic mice similar to VCD-induced follicle loss in other strains of mice.14,15 E2 supplementation of OVX and VCD mice caused FSH to drop, but not to levels observed in cycling mice. This supports that exogenous E2 was only partially effective at invoking negative feedback on pituitary FSH secretion. FSH release by the pituitary is also inhibited by ovarian-derived inhibin, a member of the transforming growth factor-? family of peptides.29 Inhibin is produced by ovarian granulosa cells,30 and, without granulosa cells, whether in OVX mice or follicle-depleted ovaries of VCD mice, the absence of inhibin was reflected in the FSH levels being reduced by half with E2 treatment but still significantly greater than in the cycling, intact mice.
The cumulative steroid pool collected throughout the study was designed to mimic the cumulative hormone exposure a woman would experience during the transition from reproductive years into the perimenopause and postmenopause periods.25,26 It has been reported that the rodent adrenal does not produce androgen.29,31 However, our results in the OVX mice, as well as those of Villablanca in castrated male mice,32 suggest an extragonadal source of androgen in mice. In our previous study, plasma androstenedione measured at a single time point on day 120 after initiation of VCD treatment was one-third the concentration in a cycling control animal.14 However, in the study reported here, androstenedione levels were measured in a pooled cumulative sample, which may explain why the level was not different between vehicle control and VCD-treated mice. E2 supplementation significantly reduced androstenedione in the VCD+E2 follicle-depleted mice compared with an increase in androstenedione seen in OVX+E2 mice. Interstitial cells of the follicle-depleted ovarian tissue produce androstenedione in culture, which is blocked by E2.33 We conclude that E2 supplementation in VCD-treated mice reduced ovarian interstitial cell production of androstenedione.
An additional endocrine difference was decreased, circulating estradiol in VCD+E2-treated mice compared with OVX+E2 mice. Estrogen is inactivated and removed by sulfoconjugation catalyzed by estrogen sulfotransferase, the expression of which in the ovary is stimulated by LH, the gonadotropin that stimulates interstitial cells.34,35 The follicle-depleted ovary enriched in interstitial cells may metabolize estradiol via estrogen sulfotransferase, resulting in less estradiol in the VCD+E2 mice compared with the OVX+E2 mice.
Together, our findings suggest that studies using the VCD-treated mouse as a model for CVD and menopause may provide an endocrine milieu more like that in perimenopausal and postmenopausal women. The VCD model would be better to assess the potential impact of therapeutic interventions in primary or secondary CVD prevention, for instance, to investigate the role of cyclooxygenase 2 inhibitors,36 a subject of recent investigations. Furthermore, using a more endocrine-relevant model will enable the performance of animal studies designed to reconcile the differences reported in clinical investigations regarding the risks or benefits of hormone supplementation.2–6,18 Another uncertainty that could be addressed using the VCD model of ovarian failure is the role of androgens in CVD. It has been assumed for decades that androgen is atherogenic and estrogen atheroprotective but, careful re-examination suggests that androgen and estrogen are atheroprotective.37–39 The VCD mouse model with sustained androgen and progesterone production from the follicle-depleted ovary provides a physiological setting in which to identify the roles of these steroids in atherosclerotic lesion development. Finally, the potential utility of this model for screening therapeutic approaches will likely facilitate drug discovery in women’s health and, more importantly, provide an early warning for contraindications of drugs prescribed for women in perimenopausal years.
Acknowledgments
This work was supported by the American Heart Association fellowship, ATVB junior investigator for women 2003 (L.P.M.); and grants ES08979, ES09246, and AG21948 (P.B.H.), center grant ES-06694 (P.B.H.), and HL55517 (C.L.B.). The authors thank Patrick J. Devine for his continued support and comments, and Patricia J. Christian, David Ruecker, Meriah Earle, Eric Cheung, Samuel Marion, and Virginia S. Patch for technical assistance. The authors wish to extend a special acknowledgement to Susan D. Airhart and Allison Kitten from the Jackson Laboratory for stimulating discussions and continued support.
References
American Heart Association web site. http://www.americanheart.org/downloadable/heart/1104938287787FS10WM05rev.pdf.
Gordon T, Kannel WB, Hjortland MC, McNamara PM. Menopause and coronary heart disease. The Framingham Study. Ann Intern Med. 1978; 89: 157–161.
Kritz-Silverstein D, Barrett-Connor E, Wingard DL. Hysterectomy, oophorectomy, and heart disease risk factors in older women. Am J Public Health. 1997; 87: 676–680.
Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J; Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. J Am Med Assoc. 2002; 288: 321–333.
Grodstein F, Manson JE, Colditz GA, Willett WC, Speizer FE, Stampfer MJ. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med. 2000; 133: 933–941.
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vittinghoff E. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. J Am Med Assoc. 1998; 280: 605–613.
Haas JS, Kaplan CP, Gerstenberger EP, Kerlikowske K. Changes in the use of postmenopausal hormone therapy after the publication of clinical trial results. Ann Intern Med. 2004; 140: 184–188.
Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, Bonds D, Brunner R, Brzyski R, Caan B, Chlebowski R, Curb D, Gass M, Hays J, Heiss G, Hendrix S, Howard BV, Hsia J, Hubbell A, Jackson R, Johnson KC, Judd H, Kotchen JM, Kuller L, LaCroix AZ, Lane D, Langer RD, Lasser N, Lewis CE, Manson J, Margolis K, Ockene J, O’Sullivan MJ, Phillips L, Prentice RL, Ritenbaugh C, Robbins J, Rossouw JE, Sarto G, Stefanick ML, Van Horn L, Wactawski-Wende J, Wallace R, Wassertheil-Smoller S; Women’s Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. J Am Med Assoc. 2004; 291: 1701–1712.
Hodgin JB, Maeda N. Minireview: estrogen and mouse models of atherosclerosis. Endocrinology. 2002; 143: 4495–4501.
Holm P. Effect of estrogen on development of atherosclerosis. A review of experimental animal studies. Dan Med Bull. 2001; 48: 146–160.
Bellino FL. Nonprimate animal models of menopause: workshop report. Menopause. 2000; 7: 14–24.
Mansfield PK, Carey M, Anderson A, Barsom SH, Koch PB. Staging the menopausal transition: data from the TREMIN Research Program on Women’s Health. Womens Health Issues. 2004; 14: 220–226.
North American Menopause Society (NAMS). Menopause Practice: A Clinician’s Guide. NAMS 2004 PDF on web site http://www.menopause.org.
Mayer L, Devine PJ, Dyer CA, Hoyer PB. The follicle-deplete ovary produces androgen. Biol Reprod. 2004; 71: 130–138.
Hooser SB, Douds DP, DeMerrel DG, Hoyer PB, Sipes IG. Long-term ovarian and gonadotropin changes in mice exposed to 4-vinylcyclohexene. Reprod Toxicol. 1994; 8: 315–323.
Marsh MM, Walker VR, Curtiss LK, Banka CL. Protection against atherosclerosis by estrogen is independent of plasma cholesterol levels in LDL receptor-deficient mice. J Lipid Res. 1999; 40: 893–900.
Kaplan JR, Manuck SB, Anthony MS, Clarkson TB. Premenopausal social status and hormone exposure predict postmenopausal atherosclerosis in female monkeys. Obstet Gynecol. 2002; 99: 381–388.
Dubey RK, Imthurn B, Zacharia LC, Jackson EK. Hormone replacement therapy and cardiovascular disease:what went wrong and where do we go from here? Hypertension. 2004; 44: 789–795.
Gambrell RD Jr. The women’s health initiative reports in perspective:facts or fallacies? Climacteric. 2004; 7: 221–224.
Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, Schwartz SM. Advanced atherosclerotic lesions in the innominate artery of the Apo E knockout mouse. Arterioscler Thromb Vasc Biol. 2000; 20: 2587–2592.
Reardon CA, Getz GS. Mouse models of atherosclerosis. Curr Opin Lipidol. 2001; 12: 167–173.
Hockings PD, Roberts T, Galloway GJ, Reid DG, Harris DA, Vidgeon-Hart M, Groot PH, Suckling KE, Benson GM Repeated three-dimensional magnetic resonance imaging of atherosclerosis development in innominate arteries of low-density lipoprotein receptor-knockout mice. Circulation. 2002; 106: 1716–1721.
Nathan L, Shi W, Dinh H, Mukherjee TK, Wang X, Lusis AJ, Chaudhuri G. Testosterone inhibits early atherogenesis by conversion to estradiol: critical role of aromatase. Proc Natl Acad Sci U S A. 2001; 98: 3589–3593.
Bernini GP, Sgro M, Moretti A, Argenio GF, Barlascini CO, Cristofani R, Salventti A. Endogenous androgens and carotid intimal-medial thickness in women. J Clin Endocrinol Metab. 1999; 84: 2008–2012.
Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab. 1996; 81: 1495–1501.
Burger HG, Dudley EC, Robertson DM, Dennerstein L. Hormonal changes in the menopause transition. Recent Prog Horm Res. 2002; 57: 257–275.
Schmidt WN, Katzenellenbogen BS. Androgen-uterine interactions: an assessment of androgen interaction with the testosterone- and estrogen-receptor systems and stimulation of uterine growth and progesterone-receptor synthesis. Mol Cell Endocrinol. 1979; 15: 91–108.
Yamasaki K, Takeyoshi M, Sawaki M, Imatanaka N, Shinoda K, Takatsuki M. Immature rat uterotrophic assay of 18 chemicals and Hershberger assay of 30 chemicals. Toxicology. 2003; 183: 93–115.
van Weerden WM, Bierings HG, van Steenbrugge GJ, de Jong FH, Schroder FH. Adrenal glands of mouse and rat do not synthesize androgens. Life Sci. 1992; 50: 857–861.
Hurwitz JM, Santoro N. Inhibins, activins, and follistatin in the aging female and male. Semin Reprod Med. 2004; 22: 209–217.
Dufau ML, Miyagawa Y, Takada S, Khanum A, Miyagawa H, Buczko E. Regulation of androgen synthesis: the late steroidogenic pathway. Steroids. 1997; 62: 128–132.
Villablanca A, Lubahn D, Shelby L, Lloyd K, Barthold S. Susceptibility to early atherosclerosis in male mice is mediated by estrogen receptor alpha. Arterioscler Thromb Vasc Biol. 2004; 1055–1061.
Magoffin DA, Erickson GF. Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium. Mol Cell Endocrinol. 1982; 28: 81–89.
Tong MH, Christenson LK, Song WC. Aberrant cholesterol transport and impaired steroidogenesis in Leydig cells lacking estrogen sulfotransferase. Endocrinology. 2004; 145: 2487–2497.
Jo M, Gieske MC, Payne CE, Wheeler-Price SE, Gieske JB, Ignatius IV, Curry TE Jr, Ko C. Development and application of a rat ovarian gene expression database. Endocrinology. 2004; 145: 5384–5396.
Egan KM, Lawson JA, Fries S, Koller B, Rader DJ, Smyth EM, Fitzgerald GA. COX-2-derived prostacyclin confers atheroprotection on female mice. Science. 2004; 306: 1954–1957.
Liu PY, Death AK, Handelsman DJ. Androgens and cardiovascular disease. Endocr Rev. 2003; 24: 313–340.
Wu FC, von Eckardstein A. Androgens and coronary artery disease. Endocr Rev. 2003; 24: 183–217.
Phillips GB. Is atherosclerotic cardiovascular disease an endocrinological disorder? The estrogen-androgen paradox. J Clin Endocrinol Metab. 2005; 90: 2708–2711.(Loretta P. Mayer; Cheryl )