Molecular Psychiatry (2013) 18, 674 -- 680 & 2013 Macmillan Publishers Limited All rights reserved 1359-4184/13 www.nature.com/mp ORIGINAL ARTICLE Involvement of estrogen receptor b in maintenance of serotonergic neurons of the dorsal raphe H Suzuki1, RPA Barros1, N Sugiyama1, V Krishnan2, BC Yaden3, H-J Kim1, M Warner1,4 and J-Å Gustafsson1,2 The serotonergic neurons of the dorsal raphe (DR) nucleus in the CNS are involved in fear, anxiety and depression. Depression and anxiety occur quite frequently in postmenopausal women, but estrogen replacement to correct these CNS disorders is at present not favored because estrogen carries with it an increased risk for breast cancer. Serotonin synthesis, release and reuptake in the DR are targets of pharmaceuticals in the treatment of depression. In the present study we have examined by immunohistochemistry, the expression of two nuclear receptors, that is, the estrogen receptors ERa and ERb. We found that ERb but not ERa is strongly expressed in the DR and there is no sex difference and no change with ageing in the number of tryptophan hydroxylase (TPH)-positive neurons in the DR of wild-type (WT) mice. However, in ovariectomized (OVX) WT and in ERbÀ/À mice, there was a marked reduction in the number of TPH-positive normal-looking neurons and a marked increase in TPH-positive spindle-shaped cells. These neuronal changes were prevented in mice 1--3 weeks (but not 10 weeks) after OVX by the selective ERb agonist, LY3201, given as continuous release pellets for 3 days. The ERb agonist had no effects on glucose homeostasis. Thus, the onset of action of the ERb agonist is rapid but there is a limited window in time after estrogen loss when the drug is useful. We conclude that, rather than estradiol, ERb agonists could be useful pharmaceuticals in maintaining functional DR neurons to treat postmenopausal depression. Molecular Psychiatry (2013) 18, 674--680; doi:10.1038/mp.2012.62; published online 5 June 2012 Keywords: LY3201; postmenopausal depression; tryptophan hydroxylase INTRODUCTION ligand, 2,3-bis(4-hydroxyphenyl)-propionitrile (DPN), is diabeto- 13 Estrogen receptor (ER) b is widely expressed in the developing genic, we have examined the effects of LY3201 on glucose mouse brain and is essential for the correct layering of the homeostasis. cortex.1--3 With immunohistochemical techniques, ERb expression has been studied in several laboratories.4 There is little or no ERb expression in adult pituitary5,6 or in the endometrium,5 whereas in MATERIALS AND METHODS the mammary gland ERb is abundant and has antiproliferative, Animals pro-differentiative effects.7 In contrast to developing mouse brain WT, ERbÀ/À and ERaÀ/Àmice were generated as previously described.14--16 where ERb is abundant, in adults, expression of ERb is confined to All mice were backcrossed to a C57BL6 background for at least 10 few areas of the brain. Owing to lack of ERb expression in the generations. Animals were housed on a 12-h light/dark cycle under pituitary, selective ERb agonists do not interrupt the hypothala- controlled temperature (20--22 1C) and fed a standard pellet diet with mo--pituitary--gonadal axis, nor do they cause growth of the water provided ad libitum. For brain collection, mice were deeply mammary gland or uterus.5,8 anesthetized with carbon dioxide and perfused manually using gravity It has been known for some time that ERa is not expressed in through left ventricle with phosphate-buffered saline followed by 4% the serotonergic neurons of the dorsal raphe (DR) but that paraformaldehyde. Brains were dissected and postfixed in the same estrogen influences the serotonergic system.9 For many years, the fixative overnight at 4 1C and were processed for 5-mm paraffin sections. effects of estrogen on the DR were thought to be indirect. This To examine glucose homeostasis, skeletal muscle and white adipose view had to be revised after the discovery of ERb in 1996 and the tissue were excised, frozen immediately for real-time quantitative PCR and finding in several laboratories of the expression of ERb in neurons blood was collected from the orbital sinus for serum metabolic panel 10,11 of the DR. E2 is equally potent on ERa and ERb, but agonists measurements. selective for ERa or ERb have been synthesized. We have obtained one ERb-selective agonist from the pharmaceutical company Eli Ovariectomy and treatment with E2 or LY3201 Lilly (LY3201) and shown that it is anxiolytic.12 At the age of 6 months, 45 WT mice were ovariectomized (OVX) bilaterally In the present study we have examined the expression of ERb in through a single dorsal midline incision across the lumbar region, making the DR of (wild-type) WT mice and the consequences to the DR both ovaries accessible. The ovary-attached fat pad was gently grasped to neurons of loss of ERb. In addition, we have investigated the lift and exteriorize the ovary. Subsequently, the periovarian sac was peeled effects of LY3201 on tryptophan hydroxylase (TPH) in the DR back over the surface of the ovary, allowing removal of the whole ovary. and, as we have previously reported that one ERb-seletive This surgery was performed at Jackson Laboratory (Sacramento, CA, USA). 1Department of Biology and Biochemistry, Center for Nuclear Receptors and Cell Signaling, University of Houston, Houston, TX, USA; 2Musculoskeletal Research, Lilly Research Laboratories, Indianapolis, IN, USA; 3Department of Biology, IUPUI, Indianapolis, IN, USA and 4Department of Biosciences and Nutrition, Karolinska Institutet, Novum, Stockholm, Sweden. Correspondence: Professor J-Å Gustafsson, Department of Biology and Biochemistry, Center for Nuclear Receptors and Cell Signaling, University of Houston, 3605 Cullen Boulevard, Houston 77204, TX, USA. E-mail: [email protected] Received 31 January 2012; revised 5 March 2012; accepted 8 March 2012; published online 5 June 2012 Estrogen receptor b in the dorsal raphe H Suzuki et al 675 1, 3 and 10 weeks after OVX, mice were randomly placed in three groups release), Innovative Research of America, Sarasota, FL, USA); and (3) five mice as follows: (1) five mice were inserted with placebo pellets; (2) five mice were were inserted with LY3201 pellets (200 mgkgÀ1 per day, Eli Lilly, Indianapolis, À1 inserted with E2 pellets (10 mgkg per day, 6.3 mg per pellet for 21-day IN, USA) subcutaneously. Hormonal treatment lasted for 3 days. a ER b ER Figure 1. Expression of estrogen receptors (ERs) in the dorsal raphe (DR) nucleus of 6-month-old male mice. Several ERa-positive cells are found in the periaqueductal gray (PAG) but not in the DR (a). Many ERb-positive neurons are found in the DR (b). The dashed line shows the border of the DR. The boxes represent higher magnification of immunoreactive ERa in the PAG, and of ERb in the lateral part of the DR. Scale bars are 100 mm in lower magnification and 20 mm in higher magnification. Figure 2. Comparison of the expression of tryptophan hydroxylase (TPH) in the DR of male/female (a, b/c, d), young/aged (a, b/c, d) and (wild- type) WT/gene-deficient (a--d/e) mice (n ¼ 5). There is neither sex nor age difference of the number of TPH-positive neurons. However, there is a marked decrease of TPH in estrogen receptor (ER)bÀ/À aged male mice. Meanwhile, there is a marked increase in TPH-positive spindle- shaped cells at the lateral part of the DR of ERbÀ/À aged male mice. **Po0.01 versus WT male, 15 months (M) of age. Scale bar is 100 mmin lower magnification and 20 mm in higher magnification. & 2013 Macmillan Publishers Limited Molecular Psychiatry (2013), 674 -- 680 Estrogen receptor b in the dorsal raphe H Suzuki et al 676 Immunohistochemistry and cell count a 1:200 dilution and incubated with avidin-biotinylated horseradish Paraffin-embedded sections were de-waxed in xylene followed by peroxidase complex (Vector laboratories, Burlingame, CA, USA) for 1 h for rehydration in graded concentrations of ethanol. Antigen retrieval was ERb staining. The slides were developed using DAB (Dako, Carpinteria, CA, achieved by placing slides in 97 1C citrate buffer (pH 6.0) for 10 min. USA) and counterstained with hematoxylin. To count the number of TPH- Endogenous peroxidase was quenched by 30 min incubation in 1% H2O2 in positive cells, six serial sections that included the DR were chosen from 50% methanol followed by blocking of unspecific protein binding with each mouse, and TPH cells were counted in the entire DR and lateral part 3% bovine serum albumin for 10 min. Sections were incubated with of the DR by two independent investigators. The location of the DR and rabbit anti-ERa (1:500, Santa Cruz Biotechnology, Santa Cruz, CA, USA), lateral part of the DR was determined as previously described.17 chicken anti-ERb (1:200 home made) or mouse anti-TPH (1:800, Sigma- Aldrich, St. Louis, MO, USA) in 3% bovine serum albumin with phosphate- buffered saline and 0.1% NP40 at room temperature overnight. The effects of LY3201 on glucose homeostasis Corresponding HRP polymer solution (Bio care medical, Concord, CA, To observe the effects of LY3201 on glucose metabolism, WT male mice USA) was added for 30 min at room temperature for ERa and TPH staining. were administered LY3201 by oral gavage (0.35 mg kgÀ1 per day) for Sections were incubated for 1 h with biotinylated secondary antibodies at 3 days. There were 27 mice in the vehicle-treated and 27 in the LY3201- Figure 3. At the age of 6 months, wild-type (WT) female mice were ovariectomized (OVX). At 1, 3, and 10 weeks (W) after OVX, mice were implanted subcutaneously with placebo, E2 or LY3201 pellets.