Age-Related Changes in Estrogen Receptor in Rat Hypothalamus: A ...

0013-7227/03/$15.00/0 Printed in U.S.A.

Endocrinology 144(9):4164 – 4171 Copyright © 2003 by The Endocrine Society doi: 10.1210/en.2003-0052

Age-Related Changes in Estrogen Receptor ␤ in Rat Hypothalamus: A Quantitative Analysis TANDRA R. CHAKRABORTY, LAURIE NG,

AND

ANDREA C. GORE

Kastor Neurobiology of Aging Laboratories, Fishberg Research Center for Neurobiology, and Brookdale Department of Geriatrics and Adult Development, Mount Sinai School of Medicine, New York, New York 10029 Although the estrogen receptor ␤ (ER␤) is a major target for actions of estrogen on the brain, little is known about its neural expression during aging, when levels and the mode of estrogen release undergo substantial changes. Therefore, in the present study we examined effects of aging and estrogen treatment on the number of cells expressing the ER␤ in female rats. Two regions relevant to reproductive function were analyzed: the anteroventral periventricular nucleus (AVPV) and the principal nucleus of the bed nucleus of the stria terminalis (pBST). The numbers of ER␤-expressing cells were quantified using an unbiased stereological approach. Female rats were used at three ages [young (3– 4 months), middle-aged (10 –12 months), and old (24 –26 months)], with or without estrogen replacement. Because the estrogen milieu impacts the func-

R

EPRODUCTIVE AGING IN female mammals is characterized by alterations in the levels and release pattern of the sex steroid hormone, estrogen. In women, estrogen concentrations undergo a precipitous decline at menopause, and the risks and benefits of estrogen replacement therapy on the reproductive tract, bone, cardiovascular system, and brain are quite controversial (1, 2). These systems are all estrogen sensitive, expressing at least one of the two known nuclear estrogen receptors (ERs) in mammals, ER␣ and ER␤. ERs are nuclear transcription factors that, upon binding with estrogen, dimerize and interact with DNA response elements to modulate the expression of specific target genes (reviewed in Ref. 3). Although estrogen levels and release patterns change with aging, little is known about the corresponding changes in ERs that mediate the response to estrogen. Nevertheless, changes in ERs would potentially impact the ability of the brain and body to respond to circulating estrogens, either endogenous or exogenous from estrogen replacement therapy. Our laboratory and others have shown age-related changes in ER␣ protein and mRNA (4 – 6). Far less is known about ER␤, the focus of the present study, which is expressed abundantly in specific brain nuclei of rats, including the anteroventral periventricular nucleus (AVPV), principal nucleus of the bed nucleus of the stria terminalis (pBST), medial amygdala, paraventricular nucleus, periventricular nucleus, and others (7–9). The ER␤ knockout mouse is subfertile and Abbreviations: AVPV, Anteroventral periventricular nucleus; AVPVl, lateral anteroventral periventricular nucleus; AVPVm, medial anteroventral periventricular nucleus; ER, estrogen receptor; pBST, principal nucleus of the bed nucleus of the stria terminalis; NMDA, Nmethyl-d-aspartate; NMDAR, N-methyl-d-aspartate receptor; NR1, NMDAR subunit 1.

tion of neurotransmitter receptors such as the N-methyl-Daspartate receptor in the brain, we also investigated the colocalization of ER␤ and the obligatory N-methyl-D-aspartate receptor subunit, NR1. We observed a significant age-related decrease in ER␤ cell number in the AVPV, but not the pBST. No significant effect of estrogen on ER␤ cell number was detected in either brain region at any age. Approximately 10% and 3% of cells expressing ER␤ also coexpressed NR1 in AVPV and pBST, respectively, and this did not differ with age or treatment. Taken together, our results demonstrate 1) there are age-related changes in ER␤ cell number that are region specific; 2) this expression is not altered by estrogen replacement; and 3) a subset of ER␤ⴚpositive cells coexpresses NR1. (Endocrinology 144: 4164 – 4171, 2003)

has other phenotypic abnormalities that suggest roles for ER␤ in mediating both reproductive and nonreproductive actions of estrogen (10, 11). To our knowledge, quantitative studies on the number of ER␤ protein-expressing cells in the brain have never been published. Because of the potential importance of this receptor in mediating effects of changing estradiol levels with aging, in the present study we first undertook an unbiased stereological analysis to quantify the number of ER␤-expressing cells in two brain regions of female rats important for reproduction, the AVPV and pBST, both of which express ER␤ in rats (9, 12), are estrogen sensitive, and are critically involved in reproductive function (13, 14). These experiments were performed in the context of aging, and in addition, we examined the effects of estrogen treatment on ER␤ cell number. Second, we quantified the number of N-methyl-d-aspartate (NMDA) receptor (NMDAR) subunit 1 (NR1)-immunoreactive cells and the colocalization of ER␤ with NR1. The coexpression of ER␤ and NR1 would indicate a potential site at which the actions of glutamate, the principal excitatory neurotransmitter in the brain (15), and those of estrogen could converge on target cells. Moreover, our present analysis on ER␤ would complement our previous work on ER␣ and its colocalization with NR1 in the hypothalamus (4, 16). Finally, we examined whether the volume of the AVPV and pBST are altered by estrogen and aging in female rats, as this would indicate more global changes in these nuclei. Materials and Methods Animals Female Sprague Dawley rats were purchased at 3– 4 months (young, n ⫽ 11), 10 –12 months (middle-aged, n ⫽ 11), or 24 –26 months (old, n ⫽ 12) of age from Harlan Sprague Dawley, Inc. (Indianapolis, IN). Rats

4164

Chakraborty et al. • Age-Related Changes in ER␤ in Rat Hypothalamus

were housed two per cage in a room with controlled temperature and were provided food and water ad libitum. The light cycle was 12 h of light and 12 h of darkness (lights on at 0700 h). All animal studies were conducted in accordance with The Guide for the Care and Use of Experimental Animals, using protocols approved by the institutional animal care and use committee at Mount Sinai School of Medicine.

Surgical procedure The rats were ovariectomized bilaterally under isoflurane anesthesia and were allowed to recover for 2–3 wk. Then rats were again anesthetized with isoflurane anesthesia and implanted sc with SILASTIC brand capsules (Dow Corning, Midland, MI) containing estrogen (10% 17␤-estradiol and 90% cholesterol; n ⫽ 6 each for young, middle-aged, and old, respectively) or vehicle (100% cholesterol; n ⫽ 5, 5, and 6 for young, middle-aged, and old, respectively). Capsules were made using SILASTIC brand laboratory tubing (inner diameter, 1.96 mm; outer diameter, 3.18 mm; Dow Corning) that was packed with estrogen or vehicle and capped with SILASTIC brand adhesive (100% silicone; Dow Corning), and allowed to dry. They were soaked in saline for at least 24 h before implantation. Young animals received an implant that was 1 cm in length, middle-aged animals received an implant that was 1.5 cm in length, and aged animals received an implant that was 2 cm in length. Different implant lengths were used for animals of different ages to account for differences in body weights. We reported in previous studies that such capsules result in circulating estradiol levels in all three age groups similar to those in young, proestrous rats (17, 18). Two days later, between 1500 –1700 h, a time chosen to represent the period of estrogen positive feedback on GnRH and gonadotropin release (19 –21), the animals were deeply anesthetized with 0.35– 0.5 ml ketamine (100 mg/ml) and 0.35– 0.5 ml xylazine (20 mg/ml). The rats were perfused initially with 3.75% acrolein and 1% paraformaldehyde in PBS (50 ml) at a rate of 50 ml/min, followed by 4% paraformaldehyde in PBS (500 ml) (22). The brains were removed and postfixed for 2– 4 h in 4% paraformaldehyde and then transferred into PBS with 0.1% sodium azide (22). Tissue sections (40 ␮m) were cut on a Vibratome (VT 1000S, Leica Instruments, Nussloch, Germany) and stored in PBS with 0.1% sodium azide.

Immunocytochemistry for stereological analyses Tissue sections were taken in a 1:2 series across the AVPV and pBST. Sections were treated with 1% sodium borohydride (30 min) and were rinsed in buffer at room temperature on a shaker (22). Then, the sections were treated to eliminate any endogenous peroxide activity (3:1, methanol/3% H2O2, 20 min at room temperature). Sections were then washed, incubated in the rabbit anti-ER␤ polyclonal antibody (1 ␮g/ml; Zymed Laboratories, South San Francisco, CA) in 10% normal goat serum and 10% normal horse serum for 3 d at 4 C on a shaker. The validation of this antibody has been described by Hrabovszky et al. (23) and Shughrue and Merchenthaler (12). The sections were then rinsed and incubated in biotinylated antirabbit IgG (1:300; Vector Laboratories, Inc., Burlingame, CA) for 1 h, followed by rinsing in PBS. Then sections were incubated in avidin-biotin-peroxidase complex (Vector Laboratories, Inc.) for 1 h, rinsed in buffer, and developed in a 3,3⬘-diaminobenzidine/peroxidase reaction. The sections were then rinsed in PBS and incubated in primary antibody to NR1 (54.1; 5 ␮g/ml dilution) in 10% normal goat serum and 10% normal horse serum for 3 d at 4 C on a shaker. This antibody has also been extensively validated (24, 25). Sections were washed and incubated in biotinylated antimouse IgG (1:300; Vector Laboratories, Inc.) for 1 h. They were then incubated in avidinbiotin-peroxidase complex (Vector Laboratories, Inc.) for 1 h. After a series of washes they were developed in VIP/peroxidase reaction (VIP substrate kit, Vector Laboratories, Inc.). Sections were rinsed, dried at room temperature, dehydrated in a graded alcohols series, stained with cresyl violet, and coverslipped with DPX (Fluka, Steinheim, Germany). Because of the large number of tissues, several immunocytochemistry runs were performed, each containing at least one animal from each of the six treatment groups.

Stereological analysis For stereological analyses each cresyl violet-counterstained region (AVPV or pBST) was outlined at low magnification (⫻10) on the live

Endocrinology, September 2003, 144(9):4164 – 4171 4165

computer image with the help of a rat brain atlas (26). The rostral-caudal borders of the AVPV were approximately anterior-posterior ⫽ 8.11– 8.39, respectively, based on the brain atlas of Swanson (26), and the AVPV was demarcated most rostrally by the caudal aspect of the organum vasculosum of the lamina terminalis, where the anterior commissure first began to cross, and most caudally by where the anterior commissure was first fully crossed, at the level of the rostral medial preoptic nucleus. The pBST was caudal to the AVPV (anterior-posterior ⫽ 8.67–9.11) (26), ranging from the level of the rostral aspect of the suprachiasmatic nucleus, the fornix, and stria terminalis through rostral parts of the paraventricular nucleus. For the AVPV, based on a report showing heterogeneity of ER␤ expression with a high density of expression in the medial-most 50 ␮m and lower density in more lateral regions of female rats (8), we subdivided the AVPV into two parts. The 50-␮m medial-most portion, including the ependymal layer of the third ventricle, was called the medial AVPV (AVPVm) and the rest of the AVPV was called the lateral AVPV (AVPVl). For the pBST the study was restricted to the posterior division, principal nucleus, a region that highly expresses ER␤ (27). A ⫻63 immersion oil, 1.4 numerical aperture objective was used to achieve optimal optical sectioning during stereological analysis (16). Following stereological principles, only animals in which a complete 1:2 series across each the AVPV and pBST were included in the analyses (4, 16, 28 –30). Several animals could not be used due to tissue damage, and the final number of rats in each group was: young vehicle, n ⫽ 5; young estrogen, n ⫽ 5; middle-aged vehicle, n ⫽ 5; middle-aged estrogen, n ⫽ 5; old vehicle, n ⫽ 6; and old estrogen, n ⫽ 6. The StereoInvestigator software (MicroBrightField, Colchester, VT) placed dissector frames using a systematic random design within each contour outlining each region on a 50 ⫻ 50-␮m2 grid, and the diaminobenzidene-labeled ER␤ nuclei, the VIP-labeled NR1 cells, and those cells double-labeled for ER␤ and NR1 were counted as three separate categories within 50 ⫻ 50-␮m optical disector frames on the x-/y-axis. The final postprocessing thickness of the sections was measured by the microcator. As the mounted section thickness was, on the average, 10 ␮m, the counting frame height was kept at 6 ␮m for all sections studied. Because the neuronal number estimates were made using the optical fractionator and did not therefore depend on a direct measurement of the volume of reference of the region considered, the shrinkage of the tissue during histological processing should not influence the precision of these estimates. Measurement of the volumes of the AVPV and pBST was performed by drawing the contour at a magnification of ⫻10 and then multiplying by the total thickness and the interslice distance. For calculations, the average postshrinkage value for the z-axis was used, although volume estimates were based on postprocessing materials that have shrunk in all three dimensions. No attempt to introduce correction factors for shrinkage was made because it probably differs in the z and x-y directions. The coefficients of error and variation of the estimates were calculated as described previously (29, 30). Quantitative analyses were performed using a computer-assisted morphometry system consisting of an Axioplan 2 photomicroscope (Zeiss, Jena, Germany) equipped with an MS-2000XYZ computercontrolled motorized stage (Applied Scientific Instrumentation, Eugene, OR), a DC 330 video camera (DAGE-MTI, Michigan City, IN), a Dell microcomputer (Austin, TX), and StereoInvestigator morphometry and stereology software (MicroBrightField, Inc., Colchester, VT) for all three age groups. Stereological methods using the optical fractionator in MicroBrightField (28) were used to estimate the total number of ER␤immunoreactive nuclei in the regions of interest. Single and double labeling were performed as described previously (16), and examples are presented in Fig. 1, M–O. Stereological microscopic analyses were performed at high power (⫻63), and in the live computer image, cells were much larger than they appear in the micrographs shown in Fig. 1 (approximately five or six cells would fill the 17-in. computer monitor). The chromogens used to label ER␤ (diaminobenzidene), NR1 (VIP), and the cresyl violet stain were all easily distinguishable at this magnification. Nuclei that were single labeled for ER␤ were visible as dark brown nuclei with no detectable cytoplasmic processes. Cytoplasmic membranes that were single labeled for NR1 had cytoplasmic light brown staining in the absence of nuclear dark brown staining; these cells were labeled with cresyl violet in the nucleus. Cells that were double labeled had light brown cytoplasms indicative of NR1 and a dark brown nuclei. In those very few cases in which there was any ambiguity in determining single or double labeling, the expression of one or both chromogens was

4166

Endocrinology, September 2003, 144(9):4164 – 4171




FIG. 2. Quantification of the total number of ER␤-immunoreactive nuclei in the AVPVm (A) and AVPVl (B) of young (3– 4 months, n ⫽ 10), middle-aged (MA; 10 –12 months, n ⫽ 10), and old (24 –26 months, n ⫽ 12) rats. u, Data for ovariectomized animals given the vehicle (cholesterol) as a control; f, data for estrogen-treated animals. In both subregions of the AVPV there was an overall significant effect of age on ER␤ cell number, with levels higher in young than in middle-aged or old rats (AVPVm, P ⬍ 0.01; AVPVl, P ⬍ 0.05). Estrogen did not alter ER␤ cell numbers in the AVPV. chol, Cholesterol-treated (vehicle) rats; MA, middle-aged rats. confirmed by direct viewing through the microscope. A suitable contrast/brightness setting that yielded a high resolution image for the cells was determined and used to produce the images. The stored images then were transferred to Adobe Photoshop. Only minor adjustments of contrast and brightness were made that in no case altered the appearance of the original materials.

Statistical analysis Differences in the number of ER␤-immunoreactive nuclei, NR1positive cells, and double-labeled cells were determined in the AVPV and pBST regions and compared between the cholesterol- and estrogentreated animals of each age group. In each case, two-way ANOVA was performed by StatView 5.0 software for Macintosh computer (variables: age and estrogen treatment). When indicated, post hoc analysis was performed using Fisher’s protected least significant difference test. Effects were considered significant at P ⬍ 0.05.

Results Expression of ER␤ in AVPV and pBST

ER␤ immunostaining was abundant and concentrated in the AVPV and pBST of female rats, as described previously (8, 9, 12). ER␤ immunoreactivity was seen at high density in cell nuclei, and NR1 immunoreactivity was concentrated in cytoplasm, probably on membrane surfaces. Figure 1 shows low power micrographs of the AVPV and pBST and their corresponding positions on a rat brain atlas (26). High power images of ER␤ and NR1 single- and double-labeled cells are shown in Fig. 1, M–O.

old animals (P ⬍ 0.01 and 0.05, respectively) for both the medial and lateral parts of the AVPV. No significant effect of estrogen (AVPVm, P ⫽ 0.38; AVPVl, P ⫽ 0.14) was found, and no interactions of age and estrogen were detected. Because results were similar in the AVPVm and AVPVl, we also determined effects of age on ER␤ cell numbers in the entire AVPV (m ⫹ l), with results similar to those seen in the two subregions (young vs. middle-aged, P ⬍ 0.005; young vs. old, P ⬍ 0.01). Effect of age and estrogen on NR1-immunoreactive cell numbers in the AVPV

The effects of age, estrogen, and their interactions on the number of NR1-immunoreactive cells were determined in the medial and lateral parts of the AVPV. In the AVPVm, relatively few NR1-immunoreactive cells were found (Fig. 3A), and no effects of age, estrogen treatment, or interactions on their numbers were detected by ANOVA. Higher numbers of NR1-positive cells were found in the AVPVl compared with the AVPVm (Fig. 3B). A significant effect of age on NR1 cell number was found in the AVPVl (P ⬍ 0.05), in which numbers of NR1-positive cells were significantly higher in young than middle-aged (P ⬍ 0.05) or old (P ⬍ 0.05) animals (Fig. 3B). No significant effect of estrogen (P ⫽ 0.12) was found, nor any interaction of age and estrogen.

Effects of age and estrogen on ER␤-immunoreactive cell numbers in AVPV

Effects of age and estrogen on numbers of ER␤ and NR1 colocalized cells in the AVPV

The effects of age, estrogen, and their interactions on the number of ER␤-immunoreactive cells were determined in the medial and lateral parts of the AVPV (Fig. 2, A and B, respectively). ANOVA indicated a significant effect of age in both subregions of the AVPV (AVPVm, P ⬍ 0.01; AVPVl, P ⬍ 0.05). Post hoc analysis demonstrated that ER␤ cell numbers were significantly greater in young than in middle-aged or

The number of cells containing both ER␤ and NR1 was determined in the medial and lateral parts of AVPV. Few colocalized cells were detected (Fig. 4, A and B), representing approximately 10% of the total number of ER␤-immunoreactive cells. Neither age nor estrogen had any effect on the number of cells expressing both ER␤ and NR1 in the AVPVm and AVPVl (Fig. 4, A and B, respectively).

FIG. 1. Photomicrographs of ER␤- and NR1-immunoreactive cells in the AVPV and pBST of representative female rats. A, B, and C are modified from a rat brain atlas (26), showing the AVPV (outlined in blue) and surrounding regions. D, E, and F are photomicrographs of rat tissues corresponding to the levels shown in A, B, and C, respectively, with the AVPV also outlined in blue (magnification, ⫻5). For analyses, the 50-␮m most medial portion was considered the AVPVm, and the rest of the AVPV was considered the AVPVl. G, H, and I are modified from a rat brain atlas, showing the pBST (outlined in blue) and surrounding areas. J, K, and L are photomicrographs corresponding to these three levels (G, H, and I, respectively), with the pBST outlined in blue (magnification, ⫻5). M, N, and O are high power photomicrographs (magnification, ⫻63), illustrating single-labeled ER␤-immunoreactive nuclei (dark brown, M), single-labeled NR1-immunoreactive cytoplasmic membranes (light brown, N), and cells in which ER␤ and NR1 are colocalized (O). Single- and double-labeled cells are indicated. All tissues were counterstained with cresyl violet.

4168



FIG. 3. Quantification of the total number of NR1-immunoreactive nuclei in the AVPVm (A) and AVPVl (B) of young (3– 4 months, n ⫽ 10), middle-aged (MA; 10 –12 months, n ⫽ 10), and old (24 –26 months, n ⫽ 12) rats. u, Data for ovariectomized animals given the vehicle (cholesterol) as a control; f, data for estrogen-treated animals. No effect of age or estrogen was seen in the AVPVm. In the AVPVl the number of NR1-immunoreactive cells was significantly greater in young than in middle-aged or old rats (P ⬍ 0.05 for both comparisons). In the AVPVl there was no effect of estrogen on NR1 cell number. chol, Cholesterol-treated (vehicle) rats; MA, middle-aged rats. FIG. 4. Quantification of the total number of ER␤ and NR1 colocalized cells in the AVPVm (A) and AVPVl (B) of young (3– 4 months; n ⫽ 10), middle-aged (MA; 10 –12 months; n ⫽ 10), and old (24 –26 months; n ⫽ 12) rats. u, Data for ovariectomized animals given the vehicle (cholesterol) as a control; f, data for estrogen-treated animals. In both subregions of the AVPV the number of ER␤/ NR1-immunoreactive colocalized cells was not affected by age or estrogen treatment. chol, Cholesterol-treated (vehicle) rats; MA, middle-aged rats.

Effects of age and estrogen on ER␤, NR1, and their colocalization in the pBST

In the pBST, no effect of age or estrogen treatment was observed in the number of ER␤ (Fig. 5A), NR1 (Fig. 5B), or ER␤ and NR1 colocalized (Fig. 5C) cells. The number of ER␤ single- and double-labeled cells was comparable to that in the AVPV, whereas the number of NR1 cells in the pBST was considerably higher than that in the AVPV. Overall, approximately 3% of ER␤ cells in the pBST coexpressed NR1. Effects of age and estrogen on the volumes of the AVPV and pBST

No effect of age or treatment was observed on the mean regional volume of the total AVPV (comprising both lateral and medial subregions; P ⫽ 0.32; Fig. 6A). In the pBST there was a significant effect of age (P ⬍ 0.05), but not estrogen (P ⫽ 0.15), on pBST volume. Post hoc analysis showed a significant decrease in pBST volume from the young to the middle-aged group (P ⬍ 0.05; Fig. 6B). Although there was a trend for volume to be lower in the old compared with the young group, this difference was not significant (P ⫽ 0.09; Fig. 6B). Discussion

In the present study we quantified the number of cells expressing the receptors ER␤, the NMDAR (NR1 subunit), and their coexpression in the AVPV and pBST. The AVPV

and pBST play important roles in reproductive physiology and behavior, and both regions express high levels of steroid hormone receptor mRNA and protein in rats (8, 9, 16, 31, 32). These regions are interconnected, as neurons arising from the pBST project to the AVPV in a sexually dimorphic manner (33, 34). In addition, the AVPV and pBST contain a high density of NMDARs (4, 14, 16, 35) that mediate effects of glutamatergic input onto target neurons. Glutamate is the principal excitatory neurotransmitter in the brain, including the hypothalamus and preoptic area (15), and its actions are involved in the maintenance and regulation of reproductive and other neuroendocrine functions (36). Thus, the AVPV and pBST are sites at which actions of estrogen and glutamate can converge to impact reproductive neuroendocrine physiology and behavior. Age-related changes in ER␤ in the AVPV

We observed that the distribution of ER␤ in the AVPV of female rats is heterogeneous, with the greatest concentration in the most medial part of the AVPV and a more diffuse expression in the lateral part of the AVPV. This observation confirms a previous report showing high levels of expression of ER␤ protein in the medial region of the AVPV in this species (8). Our study extended the analyses to a quantification of the number of ER␤ protein-expressing cells in aging animals using an unbiased stereological approach. We ob-



FIG. 5. Quantification of the total number of ER␤ (A), NR1 (B), and ER␤ and NR1 colocalized cells (C) in the pBST of young (3– 4 months; n ⫽ 10), middleaged (MA; 10 –12 months; n ⫽ 10), and old (24 –26 months; n ⫽ 12) rats. In the pBST, the numbers of ER␤-immunoreactive cells, NR1-immunoreactive cells, and colocalized cells were not significantly affected by age or estrogen treatment. u, Data for ovariectomized animals given the vehicle (cholesterol) as a control; f, data for estrogen-treated animals. chol, Cholesterol-treated (vehicle) rats; MA, middle-aged rats.

FIG. 6. Estimation of mean regional volumes of the AVPV (A, medial and lateral regions combined) and pBST (B). Data are expressed as the mean regional volume ⫻ 103 ␮m3 ⫾ SEM. No statistically significant differences were observed between any age groups or between estrogen- and cholesterol-treated animals in the AVPV. A significant effect of age was observed in the pBST between the young and MA group (P ⬍ 0.05). u, Data for ovariectomized animals given the vehicle (cholesterol) as a control; f, data for estrogen-treated animals. chol, Cholesterol-treated (vehicle) rats; MA, middle-aged rats.

served a significant decrease in the number of ER␤-immunoreactive cells during aging, with higher numbers in young than middle-aged and old rats in both the medial and lateral regions of the AVPV. Therefore, although the density of expression of ER␤ is heterogeneous, the age-related decrease in the number of ER␤-expressing cells does not differ between the medial and lateral regions of the AVPV. Our measurements of AVPV volume showed that this nucleus as a whole does not change in size during aging. Wilson et al. (6) showed a decrease in ER␤ mRNA expression in the supraoptic nucleus of middle-aged and old compared with young ovariectomized, estradiol-treated rats. Thus, the brains of female rats appear to undergo specific age-related changes, manifested at least in part by decreases in the numbers of ER␤-expressing cells in certain brain regions. AVPV and NMDARs

The role of estrogen in the brain extends to its interplay with several neurotransmitters, including glutamate, acting through its NMDAR and non-NMDARs (reviewed in Refs. 36 and 37). NMDARs are localized in many of the same brain

regions in which ERs are expressed, and the effects of NMDAR activation can be enhanced by estrogen (36, 38). Moreover, the NMDAR system itself undergoes age-related changes in the forebrain, hippocampus, and hypothalamus of old animals (18, 39 – 42). The AVPV is necessary for the estrogen-induced preovulatory GnRH/LH surge, as lesions of this region obliterate the surge and block Fos expression in ipsilateral GnRH neurons during the surge (43, 44). These functions appear to involve glutamate, as blockade of the NMDAR also reduces or attenuates the surge (38, 45), indicating an interaction between the ER and the NMDAR in the regulation of this crucial physiological function. To investigate the relationship between the expression of ER␤ and NMDAR we performed a quantitative stereological analysis on the number of NR1immunoreactive cells in the AVPV as well as the percentage of ER␤ cells that coexpress NR1. We chose the NR1 subunit because it is obligatory for the formation of a functional NMDAR (46). For cells that expressed NR1 alone (no ER␤) we found relatively low numbers in the AVPVm compared with the AVPVl, and we saw an age-related decrease that

4170


was specific to the AVPVl. We also observed low coexpression of ER␤ and NR1 in the AVPV in all age groups, with only about 10% of all ER␤-immunoreactive cells also coexpressing NR1. There was no age-related difference in the number of ER␤/NR1 double-labeled cells. Thus, the number of ER␤immunoreactive cells underwent age-related decreases in both subregions of the AVPV, NR1 underwent an age-related decrease only in the AVPVl, and the number of cells expressing both ER␤ and NR1 did not change with aging. Therefore, there is heterogeneity in the expression and regulation of molecules in the AVPV during aging. Comparison of ER␣ and ER␤ in the AVPV

We recently performed a parallel set of studies on ER␣ expression in the AVPV of ovariectomized female rats subjected to the same experimental conditions as the rats used in the present study for ER␤ analysis. We found that the number of ER␣ cells increases with age in the AVPV (4) in contrast to the age-related decrease in the number of ER␤expressing cells observed in the present study. Therefore, the age-related decrease in ER␤ cell number is not generalizable to all cell types. Moreover, our finding that the AVPV volume did not change during aging indicates that there are not global changes in cell number in this brain region. Other differences in the expression of ER␣ and ER␤ were also found in the AVPV. For example, ER␣-immunoreactive cells are more evenly distributed throughout the AVPV of the rat (4, 16), unlike ER␤, which is expressed more densely in the medial part of the AVPV (8). Approximately 88% of ER␣immunoreactive cells also coexpress NR1 (4, 16) but only about 10% of ER␤-positive cells also coexpress NR1. In the rat, estrogen levels are maintained at high levels even with advancing age, but the pattern of estrogen release changes from a cyclic to a chronic pattern of release (47). Differences in the expression of ERs in the rat brain may be a consequence of changes in the pattern of estrogen release and may represent a change in the ability of the brain to respond to estrogen. Moreover, a shift in the relative ratio of ER␣ (which increases) to ER␤ (which decreases) during aging may even be involved in the transition to acyclicity in ovarian-intact rats, although further experiments are necessary to test this hypothesis. Nevertheless, taken together, these findings indicate differential age-related regulation of the two genomic ERs, and that they are found in cells with differential glutamate sensitivity. Our results further suggest that estrogen and glutamate (the latter acting through its NMDA receptor) can interact directly within the same target cells in the hypothalamus, expressing mainly ER␣ and, to a lesser extent, ER␤. ER␤, NR1, and their coexpression in the pBST

ER␤ and NR1 were also expressed in the pBST, as reported previously (7, 9, 35, 48). The numbers of ER␤-expressing cells were similar in the pBST and AVPV, whereas NR1 expression was much higher in pBST than the AVPV. In contrast, a smaller percentage of ER␤ cells in pBST also coexpressed NR1 (⬃3%) compared with the AVPV (⬃10%). No effect of age or estrogen was observed on the numbers of ER␤immunoreactive cells, NR1 cells, or their colocalization in pBST. We did, however, find a change in pBST volume with


aging, with a significant decrease from the young to the middle-aged group. The age-related decrease in the number of ER␤ and NR1 cells occurring in the AVPV, but not the pBST, indicates that such changes are region specific. For the pBST, although we do not know whether this is due to an overall cell loss or an atrophy of specific cell types, our finding that neither ER␤ nor NR1 cell number changed in this region with aging suggests that there is not a universal cell loss in the pBST with aging, but that there may be decreases in some cell types that result in a net loss in pBST size. The expression of ER␤ in the forebrain has some notable species differences. In the mouse, unlike the rat, ER␤ may not be expressed in the AVPV, and this may be strain dependent, as indicated by preliminary observations in our laboratory (unpublished observations). In the mouse brain, ER␤ mRNA expression is lower in the rostral preoptic area and anterior hypothalamus compared with that in the rat (48). The female ER␤ knockout mouse has reduced fertility (10, 11), suggesting that although the ER␤ plays a role in reproduction, it is not absolutely obligatory for reproductive competence. The relatively lower expression of ER␤ in many preoptic and hypothalamic areas of mice compared with rats suggests potential species differences in its function. In the pBST, ER␤ is expressed abundantly in all species examined to date, including rat (7, 9, 49), mouse (48), rhesus monkey (50), and sheep (51, 52). Although ER␤ expression in the AVPV and BST of humans has not been reported, there are differences in the expression of this receptor in several hypothalamic and nonhypothalamic regions between humans and other species (53). Although we do not know whether some of these species differences in ER␤ distribution may account for or play some role in differences in reproductive physiology, it is possible that such is the case. In conclusion, we found an age-related decrease in ER␤⫺ and NR1-immunoreactive cell numbers in the AVPV, but not in the pBST. The number of cells coexpressing ER␤ and NR1 was relatively low in the AVPV and pBST (10% and 3%, respectively) and did not differ with aging. Estrogen replacement did not alter any of these parameters. Thus, our results indicate that aging significantly impacts the number of cells expressing ER␤ and NR1 in the AVPV, independently of the estrogen milieu. Acknowledgments We thank Dr. B. McEwen for suggestions and critical reading of the manuscript. We also thank Drs. J. H. Morrison, P. Hof, and T. Milner for their support during the course of this work. We acknowledge the technical help of Chet C. Sherwood. Received January 10, 2003. Accepted June 2, 2003. Address all correspondence and requests for reprints to: Andrea C. Gore, Ph.D., Pharmacology/Toxicology A1915, University of Texas, Austin, Texas 78712. E-mail: [email protected]. This work was supported by NIH Grant AG-16765 (to A.C.G.).

References 1. Wich BK, Carnes M 1995 Menopause and the aging female reproductive system. Endocrinol Metab Clin North Am 24:273–295 2. Roussouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SAA, Howard BV, Johnson KC, Kotchen JM, Ockene J 2002 Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women’s Health Initiative randomized controlled trial. JAMA 288:321–333

Chakraborty et al. • Age-Related Changes in ER␤ in Rat Hypothalamus 3. McEwen BS, Alves SE 1999 Estrogen actions in the central nervous system. Endocr Rev 20:279 –307 4. Chakraborty TR, Hof PR, Ng L, Gore AC, Stereological analysis of estrogen receptor ␣ (ER␣) expression in rat hypothalamus and preoptic area, and its regulation by aging and estrogen. J. Comp Neurol, in press 5. Funabashi T, Kimura F 1994 Effects of estrogen and estrogen receptor messenger RNA levels in young and middle-aged female rats: comparison of medial preoptic area and mediobasal hypothalamus. Acta Biol Hung 45:223–231 6. Wilson ME, Rosewell KL, Kashon ML, Shughrue PJ, Merchenthaler I, Wise PM 2002 Age differentially influences estrogen receptor-␣ (ER␣) and estrogen receptor-␤ (ER␤) gene expression in specific regions of the rat brain. Mech Age Dev 123:593– 601 7. Shughrue PJ, Komm B, Merchenthaler I 1996 The distribution of estrogen receptor-␤ mRNA in the rat hypothalamus. Steroids 61:678 – 681 8. Orikasa C, Kondo Y, Hayashi S, McEwen BS, Sakuma Y 2002 Sexually dimorphic expression of estrogen receptor ␤ in the anteroventral periventricular nucleus of the rat preoptic area: implication in luteinizing hormone surge. Proc Natl Acad Sci USA 99:3306 –3311 9. Shughrue PJ, Lane MV, Merchenthaler I 1997 Comparative distribution of estrogen receptor-␣ and -␤ mRNA in the rat central nervous system. J Comp Neurol 388:507–525 10. Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA, Smithies O 1998 Generation and reproductive phenotypes of mice lacking estrogen receptor ␤. Proc Natl Acad Sci USA 95:15677–15682 11. Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors ␣ (ER␣) and ␤ (ER␤) on mouse reproductive phenotypes. Development 127:4277– 4291 12. Shughrue PJ, Merchenthaler I 2001 Distribution of estrogen receptor ␤ immunoreactivity in the rat central nervous system. J Comp Neurol 436:64 – 81 13. Hines M, Davis FC, Coquelin A, Goy RW, Gorski RA 1985 Sexually dimorphic regions in the medial preoptic area and the bed nucleus of the stria terminalis of the guinea pig brain: a description and an investigation of their relationship to gonadal steroids in adulthood. J Neurosci 5:40 – 47 14. Gu G, Varoqueaux F, Simerly RB 1999 Hormonal regulation of glutamate receptor gene expression in the anteroventral periventricular nucleus of the hypothalamus. J Neurosci 19:3213–3222 15. van den Pol AN, Wuarin JP, Dudek FE 1990 Glutamate, the dominant excitatory transmitter in neuroendocrine regulation. Science 250:1276 –1278 16. Chakraborty TR, Ng L, Gore AC 2003 Colocalization and hormone regulation of estrogen receptor ␣ and NMDA receptor in the hypothalamus of female rats. Endocrinology 144:299 –305 17. Adams MM, Oung T, Morrison JH, Gore AC 2001 Length of post-ovariectomy interval and age, but not estrogen replacement, regulate N-methyl-d-aspartate receptor mRNA levels in the hippocampus of female rats. Exp Neurol 170:345–356 18. Gore AC, Oung T, Woller MJ 2002 Age-related changes in hypothalamic gonadotropin-releasing hormone and N-methyl-d-aspartate receptor gene expression, and their regulation by oestrogen, in the female rat. J Neuroendocrinol 14:300 –309 19. Gore AC, Roberts JL 1995 Regulation of gonadotropin-releasing hormone gene expression in the rat during the luteinizing hormone surge. Endocrinology 136:889 – 896 20. Rubin BS, Bridges RS 1989 Alterations in luteinizing hormone-releasing hormone release from the mediobasal hypothalamus of ovariectomized, steroidprimed middle-aged rats as measured by push-pull perfusion. Neuroendocrinology 49:225–232 21. Hoffman GE, Lee WS, Attardi B, Yann V, Fitzsimmons MD 1990 Luteinizing hormone-releasing hormone neurons express c-Fos antigen after steroid activation. Endocrinology 126:1736 –1741 22. Milner TA, McEwen BS, Hayashi S, Li CJ, Reagan LP, Alves SE 2001 Ultrastructural evidence that hippocampal ␣ estrogen receptors are located at extranuclear sites. J Comp Neurol 429:355–371 23. Hrabovszky E, Shughrue PJ, Merchenthaler I, Hajszan T, Carpenter CD, Liposits Z, Petersen SL 2000 Detection of estrogen receptor-␤ messenger ribonucleic acid and [125I]estrogen binding sites in luteinizing hormonereleasing hormone neurons of the rat brain. Endocrinology 141:3506 –3509 24. Siegel SJ, Brose N, Janssen WG, Gasic GP, Jahn R, Heinemann SF, Morrison JH 1994 Regional, cellular, and ultrastructural distribution of N-methyl-daspartate receptor subunit 1 in monkey hippocampus. Proc Natl Acad Sci USA 91:564 –568 25. Huntley GW, Vickers JC, Janssen W, Brose N, Heinemann SF, Morrison JH 1994 Distribution and synaptic localization of immunocytochemically identified NMDA receptor subunit proteins in sensory-motor and visual cortices of monkey and human. J Neurosci 14:3603–3619 26. Swanson LW 1998 Brain maps: structure of the rat brain. 2nd ed. Amsterdam: Elsevier 27. Laflamme N, Nappi RE, Drolet G, Labrie C, Rivest S 1998 Expression and neuropeptidergic characterization of estrogen receptors (ER␣ and ER␤) throughout the rat brain: anatomical evidence of distinct roles of each subtype. J Neurobiol 36:357–378


28. West MJ, Slomianka L, Gundersen HJ 1991 Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator. Anat Rec 231:482– 497 29. Schmitz C 1998 Variation of fractionator estimates and its prediction. Anat Embryol 198:371–397 30. Schmitz C, Hof PR 2000 Recommendations for straightforward and rigorous methods of counting neurons based on a computer simulation approach. J Chem Neuroanat 20:93–114 31. Simerly RB, Carr AM, Zee MC, Lorang D 1996 Ovarian steroid regulation of estrogen and progesterone receptor messenger ribonucleic acid in the anteroventral periventricular nucleus of the rat. J Neuroendocrinol 8:45–56 32. Simerly RB 2002 Wired for reproduction: organization and development of sexually dimorphic circuits in the mammalian forebrain. Annu Rev Neurosci 25:507–536 33. Hutton LA, Gu G, Simerly RB 1998 Development of a sexually dimorphic projection from the bed nuclei of the stria terminalis to the anteroventral periventricular nucleus in the rat. J Neurosci 18:3003–3013 34. Ibanez MA, Gu G, Simerly RB 2001 Target-dependent sexual differentiation of a limbic-hypothalamic neural pathway. J Neurosci 21:5652–5659 35. Gracy KN, Pickel VM 1995 Comparative ultrastructural localization of the NMDAR1 glutamate receptor in the rat basolateral amygdala and bed nucleus of the stria terminalis. J Comp Neurol 362:71– 85 36. Gore AC 2001 Gonadotropin-releasing hormone neurons, NMDA receptors, and their regulation by steroid hormones across the reproductive life cycle. Brain Res Brain Res Rev 37:235–248 37. Brann DW, Zamorano PL, Chorich LP, Mahesh VB 1993 Steroid hormone effects on NMDA receptor binding and NMDA receptor mRNA levels in the hypothalamus and cerebral cortex of the adult rat. Neuroendocrinology 58:666 – 672 38. Lopez FJ, Donoso AO, Negro-Vilar A 1990 Endogenous excitatory amino acid neurotransmission regulates the estradiol-induced LH surge in ovariectomized rats. Endocrinology 126:1771–1773 39. Castornia M, Ambrosini AM, Pacific L, Ramacci MT, Angelucci L 1994 Age-dependent loss of NMDA receptors in hippocampus, striatum, and frontal cortex of the rat: prevention by acetyl-l-carnitine. Neurochem Res 19:795–798 40. Muller WE, Stoll S, Scheuer K, Meichelbock A 1994 The function of the NMDA-receptor during normal brain aging. J Neural Transm 44(Suppl):145–158 41. Adams MM, Morrison JH, Gore AC 2001 N-Methyl-d-aspartate receptor mRNA levels change during reproductive senescence in the hippocampus of female rats. Exp Neurol 170:171–179 42. Bonavera JJ, Swerdloff RS, Sinha Hakim AP, Lue YH, Wang C 1998 Aging results in attenuated gonadotropin releasing hormone-luteinizing hormone axis responsiveness to glutamate receptor agonist N-methyl-d-aspartate. J Neuroendocrinol 10:93–99 43. Le WW, Wise PM, Murphy AZ, Coolen LM, Hoffman GE 2001 Parallel declines in Fos activation of the medial anteroventral periventricular nucleus and LHRH neurons in middle-aged rats. Endocrinology 142:4976 – 4982 44. Wiegand SJ, Terasawa E 1982 Discrete lesions reveal functional heterogeneity of suprachiasmatic structures in regulation of gonadotropin secretion in the female rat. Neuroendocrinology 34:395– 404 45. Urbanski HF, Ojeda SR 1990 A role for N-methyl-d-aspartate (NMDA) receptors in the control of LH secretion and initiation of female puberty. Endocrinology 126:1774 –1776 46. Monyer H, Sprengel R, Schoepfer R, Herb A, Higuchi M, Lomeli H, Burnashev N, Sakmann B, Seeburg PH 1992 Heteromeric NMDA receptors: molecular and functional distinction of subtypes. Science 256:1217–1221 47. Lu KH, Hopper BR, Vargo TM, Yen SS 1979 Chronological changes in sex steroid, gonadotropin and prolactin secretions in aging female rats displaying different reproductive states. Biol Reprod 21:193–203 48. Shughrue P, Scrimo P, Lane M, Askew R, Merchenthaler I 1997 The distribution of estrogen receptor-␤ mRNA in forebrain regions of the estrogen receptor-␣ knockout mouse. Endocrinology 138:5649 –5652 49. Isgor C, Huang GC, Akil H, Watson SJ 2002 Correlation of estrogen ␤-receptor RNA with endogenous levels of plasma estradiol and progesterone in the female rat hypothalamus, the bed nucleus of stria terminalis and the medial amygdala. Mol Brain Res 106:30 – 41 50. Gundlah C, Kohama SG, Mirkes SJ, Garyfallou VT, Urbanskii HF, Bethea CL 2000 Distribution of estrogen receptor ␤ (ER␤) mRNA in hypothalamus, midbrain and temporal lobe of spayed macaque: continued expression with hormone replacement. Mol Brain Res 76:191–204 51. Scott CJ, Tilbrook AJ, Simmons DM, Rawson JA, Chu S, Fuller PJ, Ing NH, Clarke IJ 2000 The distribution of cells containing estrogen receptor-␣ (ER␣) and ER␤ messenger ribonucleic acid in the preoptic area and hypothalamus of the sheep: comparison of males and females. Endocrinology 141:2951–2962 52. Hileman SM, Handa RJ, Jackson GL 1999 Distribution of estrogen receptor-␤ messenger ribonucleic acid in the male sheep hypothalamus. Biol Reprod 60:1279 –1284 53. Osterlund MK, Gustafsson JA, Keller E, Hurd YL 2000 Estrogen receptor ␤ (ER␤) messenger ribonucleic acid (mRNA) expression within the human forebrain: distinct distribution pattern to ER␣ mRNA. J Clin Endocrinol Metab 85:3840 –3846