PAPER Estrogen receptor ß is involved in the anorectic action of estrogen

Transcription

1 (2002) 26, ß 2002 Nature Publishing Group All rights reserved /02 $ PAPER Estrogen receptor ß is involved in the anorectic action of estrogen Y-Q Liang 1, M Akishita 2, S Kim 1, J Ako 1, M Hashimoto 1, K Iijima 1, Y Ohike 1, T Watanabe 1, N Sudoh 2, K Toba 2, M Yoshizumi 1 and Y Ouchi 1 * 1 Department of Geriatric Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; and 2 Department of Geriatric Medicine, Kyorin University School of Medicine, Tokyo, Japan OBJECTIVE: Estrogen has been implicated in feeding behavior and adiposity. This study was undertaken to elucidate the mechanism underlying the anti-obesity and anorectic action of estrogen and the role of estrogen receptor (ER) in the central nervous system. METHODS AND RESULTS: Ovariectomy in 8-week-old female Wistar rats induced hyperphagia along with an increase in body weight and abdominal fat accumulation compared to control sham-operated rats. These changes were fully reversed by subcutaneous replacement of estradiol and were abrogated by pair-feeding. Then, the effects of intracerebroventricular infusion of estradiol, alone or in combination with antisense oligodeoxynucleotides (ODN), for ER in ovariectomized rats were examined. The estradiol group showed 10 20% lower daily food intake, and after the 2-week infusion period a 14% reduction in body weight with a similar reduction in abdominal fat compared to the vehicle group. The inhibitory effect of estradiol on food intake and body weight was blocked by co-administration of ER-ß antisense ODN, whereas ER-a antisense ODN did not show any influence. CONCLUSION: These results indicate that ER-ß in the central nervous system is involved in the anorectic action of estrogen. (2002) 26, doi: =sj.ijo Keywords: estradiol; ovarian hormone; energy intake; brain; antisense oligonucleotide Introduction Menopause, which is characterized by ovarian hormone withdrawal, is associated with various diseases such as osteoporosis and atherosclerosis. 1,2 Obesity, a state frequently accompanied by multiple risk factors for atherosclerosis including type 2 diabetes, hypertension and hyperlipidemia, is also accentuated by menopause. The incidence of obesity increases with age and is higher in postmenopausal women than in age-matched premenopausal women. 3,4 Estrogen replacement therapy is shown to inhibit the increase in body weight and fat accumulation in postmenopausal women. 4 Animal studies have also demonstrated that ovariectomy increases body weight, and estrogen replacement reverses this effect. 5 8 These studies suggest that estrogen plays a role in the development of obesity. *Correspondence: Y Ouchi, Department of Geriatric Medicine, Graduate School of Medicine, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo , Japan. youchi-tky@umin.ac.jp Received 15 November 2001; revised 14 March 2002; accepted 20 March 2002 Regarding the mechanism of estrogen action on body composition, two pathways need to be considered: an anorectic action through the central nervous system and a direct action on tissue metabolism. Previous studies 6,7 have demonstrated that ovariectomy stimulates food intake, and estradiol replacement inhibits the effect. On the other hand, estrogen is reported to regulate the growth of fat and fat-free tissues, possibly by affecting protein synthesis 8,9 and energy expenditure. 10,11 Based on these findings, we investigated the anti-obesity effect of estrogen and the underlying mechanism, which is still unclear. In the present study, ovariectomy in female rats increased body weight and the weight of subcutaneous and peritoneal fat, with increased food intake. These changes were fully reversed by subcutaneous replacement of estradiol and were abrogated by pair-feeding. Then, using intracerebroventricular (ICV) infusion of estradiol, alone or in combination with antisense oligodeoxynucleotides (ODN) for the estrogen receptor (ER), we showed that estradiol attenuated food intake via ER-ß in the central nervous system.

2 1104 Materials and methods Animals Female 8-week-old Wistar rats (Nippon Bio-Supply Center, Tokyo, Japan) weighing g were used in this study. They were maintained on a 12:12 h light dark cycle (lights on 9:00 to 21:00) under controlled temperature (22 1 C). They were given standard chow (MF; 24.6% protein, 5.6% fat, 3.58 kcal=g; Nippon Bio-Supply Center) and water ad libitum, except for during pair-feeding experiments. All of the experimental procedures were approved by the Animal Research Committee of the University of Tokyo. Measurement of body weight, food intake and blood pressure Body weight of rats was measured in the steady-state using a balance (Sartorius g; Yamato, Tokyo, Japan). Food intake of each rat isolated in a cage was calculated by weighing the remaining chow with a balance (Sartorius g; Yamato). Systolic blood pressure was measured by the tail-cuff method (UR-1000=UR-5000; Ueda Inc., Tokyo, Japan). 12 Measurement of fat and serum estradiol concentration After the experimental period, the rats were anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg= kg). Abdominal subcutaneous fat was removed as the fat mass between the bilateral spina ilica anterior superior lineae. Peritoneal omental fat was removed as the whole gastrocolic omentum. These sites of fat sampling were chosen because these fat masses are easy to distinguish from surrounding tissues. The amount of fat was measured using a balance (Sartorius g, Yamato). Blood samples were collected from the abdominal aorta, and serum estradiol concentration was measured by sensitive radioimmunoassay. Western blot analysis The whole rat brain was removed and the periventricular region (approximately 2 mm thickness around the right lateral ventricle and the dorsal third ventricle) was dissected using coronal sections. This region was used based on the ODN distribution experiments described below. The brain tissues were homogenized in phosphate-buffered saline containing protease inhibitors (Complete Protease Inhibitor Cocktail; Boehringer Mannheim) at 4 C. The lysate (40 mg protein per lane) was separated on 10% SDS polyacrylamide gel, electroblotted onto nitrocellulose membrane, and immunoblotted with the antibody against rat ER-ß (donated by Dr Satoshi Inoue, The University of Tokyo) at final concentrations of 1:1000 at 4 C. Antibodies were detected with a horseradish peroxidase-linked secondary antibody using an enhanced chemiluminescence system (Amersham Pharmacia Biotech). Densitometric analysis was performed using an image analyzing software (NIH image version 1.61). Protocol I. Systemic administration of estrogen with ad libitum feeding Rats were randomly divided into three groups. Two groups were bilaterally ovariectomized (OVX) and the other underwent sham surgery (sham group) under ether anesthesia. After a 1 week recovery period, one group of OVX rats (OVX þ E2 group) received subcutaneous injection of estradiol dipropionate (20 m=kg; Teikoku Hormone Co., Tokyo, Japan) suspended in corn oil once a week. 12 The sham and OVX groups received the same amount of corn oil as vehicle once a week. The rats were killed 3 weeks after ovariectomy or sham surgery. Protocol II. Systemic administration of estrogen with pair-feeding The rat groups were Sham, OVX and OVX þ E2 as described in Protocol I. In this experiment, daily food supply was limited to 18 g per rat. This amount was determined based on the average ad libitum food consumption in sham rats. Complete food consumption was confirmed every day before food was given. The rats were sacrificed 3 weeks after ovariectomy or sham surgery. Protocol III. i.c.v. infusion of estrogen The rats were ovariectomized and simultaneously implanted with a permanent guide cannula targeted to the cerebroventricle 13 under sodium pentobarbital (100 mg=kg i.p.) anesthesia. The rat s head was held level in a stereotaxic frame (David-Kopf, Tujunga), and implanted with a cannula (Alzet Brain Infusion Kit, 3 5 mm, 28 gage; Alza Co., Palo Alto, CA, USA) in the right ventricle. An osmotic minipump (Alzet 2002, 0.5 ml=h; Alza Co.) prefilled with sterile 0.9% NaCl solution was connected to the i.c.v. cannula and implanted in the back. After a 1 week recovery period, the osmotic minipump was replaced with a new minipump filled with water-soluble 17ß-estradiol (0.25 mg=kg=week; Sigma Chemical Co.) or its vehicle (2-hydroxypropyl-ß-cyckidextrin; Sigma Chemical Co.) under ether anesthesia. The rats were killed after 2 weeks i.c.v. infusion. Protocol IV. i.c.v. infusion of ER antisense ODN Implantation of the i.c.v. cannula and initial osmotic minipump was performed as described in Protocol III. In this experiment, the osmotic minipump was replaced with a new minipump filled with ODN and water-soluble 17ß-estradiol or its vehicle. The sequence chosen as the antisense ODN target overlapped the translation initiation codon of ER-a or ER-ß. Two

3 kinds of antisense ODN (AS and AS2) and two kinds of control ODN (scrambled and sense ODN corresponding to AS) were made for ER-a and ER-ß. The sequences of ODN were searched using the Genbank database and were found to have little or no homology with any registered mrna. The sequences of the used ODN were as follows: ER-a AS, 5 0 -CAT GGT CAT GGT CAG-3 0 (from 7 6to þ 9); ER-a AS2, 5 0 -GGG TCA TGG TCA TGG-3 0 (from 7 2to þ 13) ER-a scrambled, ATC GTG GAT CGT GAC-3 0 ER-a sense, 5 0 -CTG ACC ATG ACC ATG-3 0 ER-ß AS, 5 0 -GAA TGT CAT AGC TGA-3 0 (from 7 6to þ 9) ER-ß AS2, 5 0 -TGT AGA ATG TCA TAG-3 0 (from 7 2to þ 13) ER-ß scrambled, 5 0 -AAG GTT ATC GCA AGT-3 0 ER-ß sense, 5 0 -TTC AGC TAT GAC ATT-3 0. ODN were synthesized as full phosphorothioate ODN, supplied as high-performance liquid chromatography-purified (Amersham Pharmacia Biotech) and dissolved in sterile 0.9% NaCl solution at a final concentration of 2 mg=ml (for 1 mg=h delivery). The rats were killed after 2 weeks i.c.v. infusion of ODN and water-soluble 17ß-estradiol or its vehicle. Additionally, fluorescein isothiocyanate (FITC)-labeled ER-a and ER-ß AS ODN were used to examine the diffusion of ODN in the brain. 14 Following i.c.v. infusion of estradiol and FITC-labeled ODN for 14 days, the brain was perfusionfixed with 4% paraformaldehyde and removed. Serial coronal sections (10 mm thick) of the brain were viewed under a fluorescent microscope. Statistical analysis The values in the text, tables and figures are expressed as mean s.e.m. The data were analyzed using one-factor ANOVA. If a statistically significant effect was found, Newman Keuls test was performed to isolate the difference between the groups. A P-value of less than 0.05 was considered significant. Results Effect of systemic administration of estrogen on food intake, body weight and fat accumulation As shown in Figure 1A, daily food intake dropped after surgery and then increased gradually during the experimental period. OVX rats showed an accelerated rise in food intake compared to that in sham group. Estrogen replacement to OVX, however, lowered the increase, and the food intake in OVX þ E2 finally reached the level in the sham group. In parallel, body weight gain was enhanced in OVX, which was reversed by estrogen replacement (Figure 1B). At 3 weeks after ovariectomy, total food intake was 12% higher and body weight was 15% higher in OVX than in the sham group (Table 1). In the three animal groups, there was a strong positive correlation between total food intake and body weight (r ¼ 0.748, P < 0.01). To investigate whether fat accumulation was associated with the increase in food intake and body weight, the amount of abdominal subcutaneous fat and peritoneal omental fat was measured. Both the amount of subcutaneous fat and the amount of peritoneal fat were positively correlated with food intake (r ¼ 0.254, P < 0.01 and r ¼ 0.200, P < 0.05, respectively) and body weight (r ¼ 0.448, P < 0.01 and r ¼ 0.379, P < 0.01, respectively). As shown in Table 1, both the amount of subcutaneous fat and the amount of peritoneal fat were higher in OVX than in sham group and OVX þ E2; the amount of subcutaneous fat was 24% higher and the amount of peritoneal fat was 13% higher in OVX than those in the sham group. In the second set of experiments, we examined whether body weight gain and body fat accumulation in OVX rats were attributable to the increased food intake. To do so, the rats were pair-fed with the daily food supply matched with rats from the sham group for 3 weeks. Interestingly, body weight and fat accumulation in OVX were comparable to those in sham and OVX þ E2 groups (Table 2) Figure 1 Time course of food intake (A) and body weight (B) in sham-operated (sham), ovariectomized (OVX), and ovariectomized and estrogenreplaced (OVX þ E2) rats fed ad libitum. Ovariectomy or sham operation was performed on day 0. Estrogen replacement was initiated on day 7 with subcutaneous injection of estradiol dipropionate (20 mg=kg weekly). n ¼ 8 for each group. ##P < 0.01 vs sham; *P < 0.05, **P < 0.01 vs OVX.

4 1106 Table 1 Effect of estrogen on body weight, food intake, weight of subcutaneous and peritoneal fat, blood pressure and serum estradiol concentrations in rats fed ad libitum Sham OVX OVX þ E 2 Total food intake (g=3 weeks) ** Body weight (g) ** Subcutaneous fat (g) ** Peritoneal fat (g) * Systolic blood pressure (mm Hg) Serum estradiol (pg=ml) * Values are expressed as mean s.e.m. n ¼ 8 for each group. *P < 0.05, **P < 0.01 vs sham and OVX þ E2 groups. Table 2 Effect of estrogen on body weight, and weight of subcutaneous and peritoneal fat in pair-fed rats Sham OVX OVX þ E 2 Body weight (g) Subcutaneous fat (g) Peritoneal fat (g) Values are expressed as mean s.e.m. n ¼ 8 for each group. Not significant among the groups. Effect of i.c.v. infusion of estradiol on food intake and body weight, and role of ER-ß The results of previous experiments suggested that estrogen inhibited appetite, resulting in low food intake and lean body mass. Thus, we examined the effect of centrally administered estradiol on food intake and body weight in ovariectomized rats using i.c.v. infusion of water-soluble 17ßestradiol or its vehicle. As shown in Figure 2A and B, daily food intake and body weight changes were significantly smaller in the estradiol group than in the vehicle group. After the 2-week infusion period, the estradiol group showed reductions of body weight by 14%, of subcutaneous fat by 17% and of peritoneal fat by 11% (data not shown) compared to those in the vehicle group. Serum estradiol concentration in the estradiol group was not increased compared to that in the vehicle group ( vs pg=ml, respectively), indicating that estradiol did not leak out of the central nervous system at the infusion rate we used. We next examined whether ER-a or ER-ß in the central nervous system mediated the anorexic effect of estrogen. For this purpose, i.c.v. infusion of antisense ODN for ER-a or ERß in combination with estradiol or its vehicle was performed. As shown in Figure 3A and C, the inhibitory effect of estradiol on food intake was abolished on day 5 by the coadministration of antisense ODN for ER-ß. Subsequently, however, ER-ß antisense ODN exhibited a partial antagonizing effect. The possible stimulatory effect of ER-ß antisense ODN alone on food intake was excluded by comparing it with that of ER-ß scrambled ODN or ER-a, antisense ODN alone (data not shown). Other ODN including ER-a antisense, ER-ß scrambled and ER-ß sense did not show a significant influence on the anorexic effect of estradiol. Consistent with the effects on food intake, only ER-ß antisense ODN antagonized the effect of estradiol on body weight; abrogating the effect on day 5 and subsequent partial inhibition (Figure 3B and D). To examine whether the antisense ODN inhibited the protein levels of ER-ß, the brain was removed after the 2-week infusion period and used for immunoblotting. In experiments using the whole rat brain, no significant effect of antisense ODN for ER-a or ER-ß was found (data not shown). Furthermore, experiments using FITC-labeled ODN showed that the distribution of ER-a and ER-ß antisense ODN Figure 2 Effect of i.c.v. infusion of 17ß-estradiol (E2) or vehicle on food intake (A) and body weight (B) in ovariectomized rats. At 1 week after ovariectomy and implantation of an i.c.v. cannula, an osmotic minipump was connected to infuse water-soluble E2 or vehicle for 2 weeks. Day 0 indicates the day when i.c.v. infusion of E2 or vehicle was initiated. n ¼ 10 for each group. **P < 0.01 vs vehicle.

5 1107 Figure 3 Effect of i.c.v. infusion of antisense oligodeoxynucleotides (ODN) for estrogen receptor (ER) on food intake (A, C) and body weight (B, D) in ovariectomized rats. At 1 week after ovariectomy and implantation of an i.c.v. cannula, an osmotic minipump was connected to infuse water-soluble 17ßestradiol (E2) or vehicle in combination with ODN for 2 weeks. Two kinds of antisense ODN (AS and AS2) for ER-a and ER-ß, and control scrambled (scrmb) and sense ODN for AS were used. Day 0 indicates the day when i.c.v. infusion of E2 or vehicle with ODN was initiated. Food intake (C) and body weight (D) on day 5 are shown as bar graphs. *P < 0.05, **P < 0.01 vs scrmb, sense and ER-a AS and AS2. #P < 0.05, ##P < 0.01 vs all other groups including ER-ß AS. was localized in the brain region around the right lateral ventricle (infusion side) and the dorsal third ventricle (data not shown). Finally, therefore, the periventricular region was dissected and subjected to Western blot analysis for ER-ß. As shown in Figure 4, ER-ß antisense ODN, but not ER-a antisense ODN, significantly inhibited the expression of ER-ß protein. Discussion In this series of experiments, we confirmed the previously reported results 7,8,15 18 that systemic or intracerebral administration of estrogen exerted an anorectic action and inhibited body weight and fat accumulation in ovariectomized rats. The novel finding of this study is that i.c.v. infusion of antisense ODN for ER-ß antagonized the anorectic effect of estrogen. The result suggests that ER-ß plays a role in the anorectic action of estrogen. The first set of experiments demonstrated that ovariectomy enhanced the body weight gain with increased abdominal fat accumulation in rats allowed free access to chow. This was fully reversed by estrogen replacement of a physiologic dose, as was shown in previous reports. 7,8 Accumulation of visceral fat is known to play more important roles than subcutaneous fat in diseases such as type 2 diabetes, hyperlipidemia and hypertension. 19 Since estrogen is reported to decrease visceral fat rather than subcutaneous fat in postmenopausal women, 4 we examined the amount of both visceral (peritoneal) and subcutaneous fat in the rats. However, estrogen inhibited the accumulation of peritoneal fat and subcutaneous fat in parallel. This negative result might arise from the short duration of our experiments, as Zamboni et al 20 reported that regain of visceral fat after very-low-energy diet did not differ between pre- and postmenopausal women. Obesity is attributable to an imbalance between energy intake and expenditure. In fact, ovariectomy increased and estrogen replacement decreased food intake to normal with a close correlation with body weight and abdominal fat in this study. Consequently, in the second set of experiments, we examined whether pair-feeding could affect the anti-obesity effect of estrogen. Interestingly, by matching daily energy intake, the difference in body weight and fat accumulation among the three animal groups disappeared. This result is

6 1108 Figure 4 Effect of i.c.v. infusion of oligodeoxynucleotides (ODN) on protein levels of estrogen receptor (ER)-ß in the brain. After 2-week i.c.v. infusion of water-soluble 17ß-estradiol and ODN, the periventricular region of the brain was dissected and used for Western blot analysis of ER-ß. Antisense ODN (AS) for ER-a and ER-ß, and control scrambled ODN (scrmb) for AS were used. Ovaries obtained from intact female rats served as the control. Representative blots (upper panels) and densitometric analyses (lower panels) are shown. *P < contradictory to the previous pair-feeding studies where keeping food consumption constant did not prevent the increase in body fat in ovariectomized rats. 7,21 The different results might arise from the timing of ovariectomy and=or the duration of the experiments. Clearly, estrogen can exert the influence on energy expenditure as shown in many studies, 7,10,11,21 but our results indicate that the anorectic action plays an important role in the anti-obesity effects of estrogen in our experimental condition. i.c.v. infusion of estradiol verified that estrogen directly modulates the central nervous system to suppress hyperphagia in ovariectomized rats. Although i.c.v. cannulation without ovariectomy was not examined, daily food intake and body weight gain for the last 2 weeks were comparable between sham, OVX þ E2 and OVX þ i.c.v. E2 groups, and between OVX and OVX þ i.c.v. vehicle groups (see Figures 1 and 2). These data suggest that i.c.v. estradiol fully reversed the ovariectomy-induced increase in food intake and body weight. Possible targets for this action include the ventromedial hypothalamus (VMH), paraventricular nucleus (PVN), arcuate nucleus (ARC) and medial preoptic area (MPOA), because these regions of the hypothalamus are important in feeding behavior. VMH-lesioned rats were less responsive to estradiol replacement compared with control ovariectomized rats. 15 In contrast, PVN lesion 16 and microinjection 17 experiments suggest that the PVN is involved in the estrogen action on energy intake. It is also reported that microinjection of estradiol into the MPOA inhibited food intake in ovariectomized rats. 18 Taken together with the wide distribution of ER in the brain, multiple brain sites are considered to be involved in the anorectic effect of estrogen. Moreover, the effect of microinjection persists for less than half a day. 18 Therefore, in order to examine the chronic effects of estrogen and ER antisense ODN, we performed i.c.v.. infusion rather than microinjection into the nucleus of interest. It took 5 days to see significant inhibition of food intake along with body weight by i.c.v. infusion of estradiol. This is probably because the volume of the infusion cannula was 40 ml, so the solution in the osmotic minipump required 80 h to reach the brain. Estradiol concentrations in cerebrospinal fluid were up to 17 pg=ml ( pg=ml; mean s.e.m., n ¼ 5) after i.c.v. estradiol infusion. Therefore, the estradiol dose we used was a little above the physiologic range, as compared to the estradiol concentrations of pg=ml in cerebrospinal fluid of the control rats (sham, n ¼ 4). Those in ovariectomized rats were under or equal to the detection limit (1.0 pg=ml) whereas those were pg=ml in ovariectomized rats with systemic estradiol replacement (OVX þ E2, n ¼ 4). Accordingly, subcutaneous replacement of estradiol reversed the estradiol concentrations in cerebrospinal fluid. The signaling mediating the anorectic action of estrogen is unclear. Previous reports suggest the role of neuropeptide Y, 25 corticotropin-releasing factor 18 or leptin. 26 In our preliminary experiments, however, the level of neuropeptide Y mrna in the brain assayed by Northern blotting was not significantly changed by i.c.v. infusion of estradiol (data not shown). Precise studies are needed to address this issue by considering the localization of neuropeptide Y in the brain 27 and the influence of fat mass on leptin production. Both ER-a and ER-ß are reported to be widely distributed in the central nervous system In the hypothalamus, both subtypes exist in the MPOA and ARC, whereas ER-a is predominant in the VMH and ER-ß in the PVN. These regions play important roles in feeding as well as in reproduction, sexual behavior and thermoregulation. Both receptors also appear to exist in the amygdala and hippocampus, where they may be involved in short-term memory and emotion. ER-ß is also present in the cerebellum and cortical regions. To examine which subtype of ER mediates the anorectic action of estrogen, we infused antisense ODN for ER-a or ER-ß together with estradiol. Antisense ODN for ERß-, but not for ER-a, inhibited the effect of estradiol on food intake and body weight. Western blot analysis demonstrated that ER-ß antisense ODN decreased the protein expression of ER-ß in the periventricular region. A large number of studies 28 successfully investigated the neuroendocrine and behavioral effects of intracerebral administration of antisense ODN including antisense ODN for ER-a. 29,30 In the present study, specific effects of the antisense ODN for ER-ß were confirmed using two different sequences of antisense ODN and control sense and scrambled ODN. This was further supported by the results of immunoblotting. Accordingly, it can be concluded that ER-ß is involved, at least in part, in the anorectic action of estrogen. The role of ER-a in the anorectic action of estrogen remains uncertain. Based on the null effect on food intake and body

7 weight, it is possible that the two kinds of ER-a antisense ODN we designed were not sufficiently potent to exert their actions, although one of them (ER-a AS) is identical to the ODN that was effective in previous i.c.v. experiments. 29,30 The finding in ER-a knockout mice provides some information on this point. In ER-a knockout mice, energy expenditure was significantly reduced compared to that in wild-type mice, whereas energy intake was unchanged, 11 suggesting that ER-a is not indispensable for the anorectic action of estrogen but is critical to energy expenditure. By comparing ER-a, ER-ß and double knockout mice in combination with antisense ODN, differential roles of ER-a and ER-ß in energy balance can be investigated in the future. Furthermore, administration of ER antisense ODN in normal rats could make the role of endogenous estrogen more clear. In summary, we have shown that energy intake plays an important role in the anti-obesity effect of estrogen in a rat model of ovariectomy-induced adiposity, and that ERß in the brain is involved in the anorectic action of estrogen. These findings will add new information to the understanding of the role of estrogen in regulating obesity and metabolism. Acknowledgements This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Grant-in-Aid for Scientific Research no , no , no ) and from the Ministry of Health, Labor and Welfare of Japan (H10-Choju-007). References 1 The North American Menopause Society. A decision tree for the use of estrogen replacement therapy or hormone replacement therapy in postmenopausal women: consensus opinion of The North American Menopause Society. Menopause 2000; 7: Genazzani AR, Gambacciani M. HRT in the third millennium. Maturitas 2001; 38(Suppl 1): S Svendsen OL, Hassager C, Christiansen C. Age- and menopauseassociated variations in body composition and fat distribution in healthy women as measured by dual-energy X-ray absorptiometry. Metabolism 1995; 44: Tchernof A, Poehlman ET, Despres JP. Body fat distribution, the menopause transition, and hormone replacement therapy. Diabetes Metab 2000; 26: Wade GN, Gray JM, Bartness TJ. Gonadal influences on adiposity. Int J Obes 1985; 9: Donohoe TP, Stevens R. Effects of ovariectomy, estrogen treatment and CI-628 on food intake and body weight in female rats treated neonatally with gonadal hormones. Physiol Behav 1983; 31: Roy EJ, Wade GN. Role of food intake in estradiol-induced body weight changes in female rats. Horm Behav 1977; 8: Ramirez I. Estradiol-induced changes in lipoprotein lipase, eating, and body weight in rats. Am J Physiol 1981; 240: E Price TM, O Brien SN, Welter BH, George R, Anandjiwala J, Kilgore M. Estrogen regulation of adipose tissue lipoprotein lipase possible mechanism of body fat distribution. Am J Obstet Gynecol 1998; 178: Richard D. Effects of ovarian hormones on energy balance and brown adipose tissue thermogenesis. Am J Physiol 1986; 250: R Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS. Increased adipose tissue in male and female estrogen receptoralpha knockout mice. Proc Natl Acad Sci USA 2000; 97: Akishita M, Ouchi Y, Miyoshi H et al. Estrogen inhibits cuffinduced intimal thickening of rat femoral artery: effects on migration and proliferation of vascular smooth muscle cells. Atherosclerosis 1997; 130: Liang J, Toba K, Ouchi Y et al. Central vasopressin is required for the complete development of deoxycorticosterone-salt hypertension in rats with hereditary diabetes insipidus. J Auton Nerv Syst 1997; 62: Whitesell L, Geselowitz D, Chavany C et al. Stability, clearance, and disposition of intraventricularly administered oligodeoxynucleotides: implications for therapeutic application within the central nervous system. Proc Natl Acad Sci USA 1993; 90: Beatty WW, O Briant DA, Vilberg TR. Effects of ovariectomy and estradiol injections on food intake and body weight in rats with ventromedial hypothalamic lesions. Pharmac Biochem Behav 1975; 3: Butera PC, Willard DM, Raymond SA. Effects of PVN lesions on the responsiveness of female rats to estradiol. Brain Res 1992; 576: Butera PC, Beikirch RJ. Central implants of diluted estradiol: independent effects on ingestive and reproductive behaviors of ovariectomized rats. Brain Res 1989; 491: Dagnault A, Richard D. Involvement of the medial preoptic area in the anorectic action of estrogens. Am J Physiol 1997; 272: R Wajchenberg BL. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev 2000; 21: Zamboni M, Armellini F, Turcato E et al. Effect of regain of body weight on regional body fat distribution: comparison between pre- and postmenopausal obese women. Obes Res 1996; 4: Hervey E, Hervey GR. Energy storage in female rats treated with progesterone in the absence of increased intake of food. J Physiol 1969; 200: 118P 119P. 22 Shughrue PJ, Lane MV, Merchenthaler I. Comparative distribution of estrogen receptor-alpha and -beta mrna in the rat central nervous system. J Comp Neurol 1997; 388: Osterlund M, Kuiper GG, Gustafsson JA, Hurd YL. Differential distribution and regulation of estrogen receptor-alpha and -beta mrna within the female rat brain. Brain Res Mol Brain Res 1998; 54: Shughrue PJ, Merchenthaler I. Estrogen is more than just a sex hormone : novel sites for estrogen action in the hippocampus and cerebral cortex. Front Neuroendocrinol 2000; 21: Shimizu H, Ohtani K, Kato Y, Tanaka Y, Mori M. Withdrawal of estrogen increases hypothalamic neuropeptide Y (NPY) mrna expression in ovariectomized obese rat. Neurosci Lett 1996; 204: Machinal F, Dieudonne MN, Leneveu MC, Pecquery R, Giudicelli Y. In vivo and in vitro ob gene expression and leptin secretion in rat adipocytes: evidence for a regional specific regulation by sex steroid hormones. Endocrinology 1999; 140: Morris BJ. Neuronal localisation of neuropeptide Y gene expression in rat brain. J Comp Neurol 1989; 290: Landgraf R, Naruo T, Vecsernyes M, Neumann I. Neuroendocrine and behavioral effects of antisense oligonucleotides. Eur J Endocrinol 1997; 137: McCarthy MM, Schlenker EH, Pfaff DW. Enduring consequences of neonatal treatment with antisense oligodeoxynucleotides to estrogen receptor messenger ribonucleic acid on sexual differentiation of rat brain. Endocrinology 1993; 133: Apostolakis EM, Garai J, Lohmann JE, Clark JH, O Malley BW. Epidermal growth factor activates reproductive behavior independent of ovarian steroids in female rodents. Mol Endocrinol 2000; 14: