Psychoneuroendocrinology (2015) 55, 81—93

Available online at www.sciencedirect.com

ScienceDirect

j ourna l h om epa ge : www.elsev ier .com/ locate /psyneuen

Serotonin 2C receptor contributes togender differences in stress-inducedhypophagia in aged mice

Chihiro Yamadaa, Chiharu Sadakanea, Miwa Nahataa,Yayoi Saegusaa, Koji Nakagawab, Naoto Okubob,Shunsuke Ohnishi c, Tomohisa Hattori a, Hiroshi Takedab,c,∗

a Tsumura Research Laboratories, Tsumura & Co., 3586 Yoshiwara, Ami-machi, Inashiki-gun,Ibaraki 300-1192, Japanb Pathophysiology and Therapeutics, Hokkaido University Faculty of Pharmaceutical Sciences, N12 W6,Kita-ku, Sapporo, Hokkaido 060-0812, Japanc Gastroenterology and Hepatology, Hokkaido University Graduate School of Medicine, N15, W7, Kita-ku,Sapporo, Hokkaido 060-8638, Japan

Received 9 August 2014; received in revised form 21 January 2015; accepted 10 February 2015

KEYWORDSAged;Gender differences;Novelty stress;Food intake;5-HT2CR;Rikkunshito

Summary The combination of depression and anorexia may influence morbid-ity and progressive physical disability in the elderly. Gender differences exist inhypothalamic—pituitary—adrenal axis activation following stress exposure. The objectiveof this study was to investigate gender differences in feeding behavior under novelty stress inaged mice. Food intake measurement, immunohistochemical assessment, and mRNA expressionanalysis were conducted to investigate the role of serotonin 2C receptor (5-HT2CR) and itsrelationship with ghrelin in stress-induced suppression of feeding behavior in aged mice.

After exposure to novelty stress, a 21-fold increase in plasma corticosterone and remarkable suppression of food intake were observed in aged male mice. Furthermore, a 5-HT2CR agonistsuppressed food intake in aged male mice. Novelty stress induced a 7-fold increase in 5-HT2CRand c-Fos co-expressing cells in the paraventricular nucleus of the hypothalamus in aged malemice but caused no change in aged female mice. Plasma acylated ghrelin levels decreased instressed aged male mice and administration of the 5-HT2CR antagonist inhibited this decrease.

∗ Corresponding author at: Pathophysiology and Therapeutics, Hokkaido University Faculty of Pharmaceutical Sciences, N12 W6, Kita-ku,Sapporo, Hokkaido 060-0812, Japan. Tel.: +81 11 706 3746; fax: +81 11 706 4978.

E-mail address: h takeda@pharm.hokudai.ac.jp (H. Takeda).

http://dx.doi.org/10.1016/j.psyneuen.2015.02.0060306-4530/© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

82 C. Yamada et al.

The 5-HT2CR antagonist also reversed the suppression of food intake in estrogen receptor �

agonist-treated aged male mice. Therefore, conspicuously suppressed feeding behavior in nov-elty stress-exposed aged male mice may be mediated by 5-HT2CR hypersensitivity, leading tohypoghrelinemia. The hypersensitivity may partly be due to estrogen receptor activation in agedmale mice.© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC

ative

1

Tid2asbadR

hfWm(Hfb2pFfTiit

posearHh2uHmtm

ev(tEr

tstfifaHdamifb

2

2

MoJoah(TgT

2pl

Tds(ogfuswere transferred to separate cages (1 mouse/cage; cagesize = W136 × D208 × H115 mm) after 18 h of fasting. Theremaining group-housed mice served as controls. Afterdecapitation without anesthesia, blood was collected from

BY-NC-ND license (http://cre

. Introduction

he number of elderly people with anxiety and depressions increasing and the effective treatment of these disor-ers remains a clinical challenge (Diefenbach and Goethe,006). Adverse life events are associated with the onsetnd persistence of depression (Paykel, 1994), and depres-ion is characterized by feeding abnormality (e.g., anorexia,ulimia). In particular, the combination of depression andnorexia may influence morbidity and progressive physicalisability in the elderly (Rowe and Kahn, 1987; Seeman andobbins, 1994).

Gender differences exist inypothalamic—pituitary—adrenal (HPA) axis activationollowing stress exposure (Handa et al., 2012; Handa andeiser, 2014; Steenbergen et al., 1990). The incidence ofajor depression (Kornstein, 1997) and anorexia nervosa

Ehlert et al., 2001) is higher in women. In young rodents,PA axis activation due to stress is reportedly greater in

emales (Iwasaki-Sekino et al., 2009) and is controlledy sex hormones (Handa et al., 2012; Handa and Weiser,014). On the other hand, the anorexia of aging is moreronounced in males than in females (Morley, 1993, 2010).urthermore, late-life depression in males is a risk factoror mortality (Diniz et al., 2014; Dombrovski et al., 2008).herefore, gender differences may have a considerable

mpact on emotional and/or feeding behavior in the elderlyn a manner different than that in young people. However,his possibility has not yet been tested.

Corticotropin-releasing factor (CRF) and serotonin (5-HT)lay important roles in stress responses and the regulationf feeding behavior. CRF neuron activation during stressuppresses feeding behavior (Hotta et al., 1999; Zorrillat al., 2003). CRF production and neuron activation maylso be mediated by 5-HT. Acute 5-HT depletion has beeneported to reduce anxiety (Bechtholt et al., 2007). The 5-T2C receptor (5-HT2CR), localized on CRF neurons of theypothalamus, stimulates anxiety (de Mello Cruz et al.,005; Gatch, 2003; Heisler et al., 2007) and negatively reg-lates food intake (Dryden et al., 1996; Hayashi et al., 2005;eisler et al., 2002; Nonogaki, 2008). Increased 5-HT2CRRNA expression in the paraventricular nucleus (PVN) of

he hypothalamus has been observed in stress-exposed agedice (Nahata et al., 2013).Estradiol-induced effects on stress may be controlled by

strogen receptor (ER) � and �. ER� decreases food intakeia the HPA axis and cholecystokinin signal amplificationAsarian and Geary, 2013), whereas ER� negatively con-

rols the HPA axis (Handa et al., 2012). However, whetherRs influence gender differences in stress-loaded animalsemains unclear.

ts

commons.org/licenses/by-nc-nd/4.0/).

The aim of the current study was to investigate whetherhere were gender differences in novelty stress-induceduppression of feeding behavior in aged mice. To clarifyhe mechanism underlying these differences, we identi-ed the effects of 5-HT2CR agonists and antagonists oneeding behavior, emphasizing on the role of 5-HT2CR. Welso conducted an immunohistochemical assessment of 5-T2CR activation in stress-related brain areas. Next, weetermined whether decreased plasma ghrelin levels medi-ted the marked suppression of feeding behavior in agedale mice. Finally, we evaluated the susceptibility of ERs

n aged male mice as a factor underlying gender dif-erences in novelty stress-induced suppression of feedingehavior.

. Materials and methods

.1. Animals

ale and female C57BL/6 mice aged 6 weeks and 79 weeksr older were purchased from Charles River Laboratoriesapan (Tokyo, Japan). We used elderly mice before thenset of aging cachexia. Before the experiments, mice werecclimated in cages (5 mice/cage) in a temperature- andumidity-controlled room under a 12-h light/dark cyclelights on at 07:00 h) with free access to food and water.his study was approved by and conducted according to theuidelines of the experimental animal ethics committees ofsumura & Co. (Ibaraki, Japan; permit no.: 09-115).

.2. Effects of exposure to novelty stress onlasma corticosterone levels, plasma ghrelin

evels, and food intake in young and aged mice

he novelty-induced hypophagia test can evaluate theegree of anxiety or depression based on the suppres-ion of food intake after exposure to a novel environmentDulawa and Hen, 2005). Tests were conducted as previ-usly described (Nahata et al., 2013). Mice were housed inroups (5 mice/cage; cage size = W230 × D310 × H155 mm)or 7 days before the experiment; cages were not changedntil the experiment was completed. To induce noveltytress, some of the group-housed mice (5 mice/cage)

he different animal groups at 3 and/or 6 h after expo-ure to novelty stress (n = 7—8). Plasma corticosterone

tres

tas

lPhSlstp(aacm

e(icg

2

BscptTHwlaCbUbfsnsbcBantac3

Serotonin 2C receptor contributes to gender differences in s

and ghrelin levels were determined using the corticoste-rone EIA kit (Enzo Life Sciences, Plymouth Meeting, PA,USA) and the Active-Ghrelin ELISA kit (Mitsubishi Chem-ical Medicine, Tokyo, Japan), respectively. Samples werecollected between 13:00 h and 15:00 h to avoid diurnalvariations.

Food intake was evaluated at 1, 6, and 24 h after expo-sure to novelty stress (n = 8—10). To evaluate food intake permouse, mice housed individually for 7 days before the exper-iment served as the control group. We confirmed that therewas no difference in the mean food intake between theseindividually housed and group-housed mice, calculated bydividing food intake for each cage by the number of mice(Nahata et al., 2013). We also confirmed that plasma corti-costerone levels in individually housed mice were similar tothose in group-housed mice (data not shown), as previouslyreported (Arndt et al., 2009).

2.3. Total RNA extraction and reversetranscription polymerase chain reaction

The hypothalamus was rapidly removed from each fastingmouse (n = 8) and immediately frozen in a tube main-tained on dry ice. Isolated tissue homogenization and totalRNA extraction were performed using the RNeasy Univer-sal Tissue kit (Qiagen, Valencia, CA, USA). Diluted totalRNA (100 ng/�L) was incubated at 70 ◦C for 5 min andthen cooled on ice. Total RNA (1000 ng) was reverse tran-scribed using the TaqMan Reverse Transcription Reagents kit(Applied Biosystems, Foster City, CA, USA). Furthermore,quantitative PCR assays were performed using the TaqManGene Expression Master Mix (Applied Biosystems) and Taq-Man gene-specific primer/probes (Rps29, Mm02342448 gH;Cyp19a1, Mm00484049 m1; Esr1, Mm00433149 m1; andEsr2, Mm00599821 m1; Applied Biosystems) on a Prism7900HT Sequence Detection System (Applied Biosystems).mRNA expression versus a housekeeping gene (ribosomalprotein S29) was calculated using the ��Ct method.

2.4. Effects of test drugs on food intake andplasma ghrelin

Young and aged mice were administered subcutaneous (SC)injections of the 5-HT2CR agonist CP-809101 (Sigma-Aldrich,St. Louis, MO, USA) at doses of 0.1, 1.0, and 3.0 mg/kg,and the cumulative food intake was measured at 1 and3 h after drug administration. To investigate the effects of5-HT2CR antagonists on food intake in aged mice, the selec-tive 5-HT2CR antagonists SB242084 [6 mg/kg, (Hayashi et al.,2005; Kennett et al., 1997)] and rikkunshito [RKT; Tsumura& Co., Tokyo, Japan; 1000 mg/kg, (Takeda et al., 2008)]were suspended in distilled water and orally administeredby gavage (10 mL/kg) to the 18 h-fasted mice immediately

after they were exposed to novelty stress (n = 5—8). More-over, the cumulative food intake was determined 6 and 24 hafter exposure to novelty stress. We have previously con-firmed that oral administration of SB242084 (6 mg/kg) andRKT (1000 mg/kg) to young or aged mice does not affect foodintake (Nahata et al., 2013).

Idpnnt

s-induced hypophagia in aged mice 83

Acylated ghrelin was then intraperitoneally administeredo the mice immediately after exposure to novelty stress,nd food intake was then determined at 1, 3, and 6 h aftertress loading.

We investigated the effect of the aromatase inhibitoretrozole [Tocris Bioscience, Ellisville, MO, USA; 2.5 mg/kg,O (Meng et al., 2011)] and the ER� agonist 2,3-bis(4-ydroxyphenyl)-propionitrile [DPN; Sigma-Aldrich; 8 mg/kg,C (Shin et al., 2013)] on the decrease in 6-h cumu-ative food intake in aged mice exposed to noveltytress, following the procedure described above. Next,he ER� agonist 1,3,5-tris(4-hydroxyphenyl)-4-propyl-1H-yrazole [(PPT; Sigma-Aldrich; 0.63 and/or 1.25 mg/kg, SCSantollo et al., 2007)] was administered to young andged mice. SB242084 (6 mg/kg, PO) was also immediatelydministered after PPT (1.25 mg/kg) treatment, and theumulative food intake in young and aged mice was deter-ined at 6 h.To clarify the role of 5-HT2CR on plasma ghrelin lev-

ls, oral administration of SB242084 (6 mg/kg) or RKT1000 mg/kg) to 18-h fasted aged male mice was performedmmediately after exposure to novelty stress, and blood wasollected at 3 h after this exposure to determine the plasmahrelin levels as described above.

.5. Immunohistochemistry

rains were removed from mice after exposure to noveltytress for 6 h in an ad libitum condition, and immunohisto-hemical analysis was performed. In brief, after transcardialerfusion with a tissue fixative (Genostaff, Tokyo, Japan),he brains were removed and embedded in paraffin blocks.he cross sections (6 �m) of brains were exposed to 5-T2CR and c-Fos. For the first staining step, tissue sectionsere treated with 0.3% H2O2 in methanol for 30 min, fol-

owed by incubation with Protein Block (Genostaff) and thevidin/biotin blocking kit (Vector Laboratories, Burlingame,A, USA). The sections were incubated with anti-5HT2CR rab-it polyclonal antibody (Novus Biologicals, Littleton, CO,SA) at 4 ◦C overnight. They were then incubated withiotin-conjugated goat anti-rabbit Ig (Dako, Tokyo, Japan),ollowed by the addition of alkaline phosphatase-conjugatedtreptavidin (Nichirei Biosciences, Tokyo, Japan). Sig-als from the sections were detected using NBT/BCIPolution (Sigma-Aldrich). For the second staining step, anti-ody dissociation was performed by heat treatment withitrate buffer. The sections were incubated with Proteinlock and the avidin/biotin blocking kit and then withnti-cFos rabbit polyclonal antibody (Santa Cruz Biotech-ology, Dallas, TX, USA) at 4 ◦C overnight. Furthermore,he sections were incubated with biotin-conjugated goatnti-rabbit Ig (Dako), followed by the addition of peroxidase-onjugated streptavidin. Their signals were detected using,3′-diaminobenzidine (Sigma-Aldrich). We used anti-rabbitg (Dako) for negative controls. The numbers of clearlyetected c-Fos-positive (blue nuclei, Fig. 4A left), 5-HT2CR-

ositive (brown cytoplasm, Fig. 4A center), and co-positiveerve cells (Fig. 4A right) in the PVN, amygdala, and arcuateucleus (ARC) were manually counted, and the averages ofhe 2 sections per mouse were calculated.

8

2

TtbowSfn

3

3c

PaTsppuam(ntif

saaewfF2pp[FooliFtpsFm

3i

Syi

Iimha8ao

3de

Cec((ft

3ce

CnoaTiuHsatpnIc[san[ectceaenaF[

4

.6. Statistical analyses

wo-way factorial analysis of variance (ANOVA) followed byhe Tukey—Kramer post hoc test was used for the comparisonetween young and aged mice. Data from the experimentsn aged mice were analyzed using Student’s t-test and one-ay analysis of variance, followed by Dunnett’s test orteel’s test. Data are presented as mean ± standard erroror each group, and p < 0.05 was considered statistically sig-ificant.

. Results

.1. Effects of novelty stress exposure onorticosterone secretion and food intake

lasma corticosterone levels were remarkably increased inged male mice 6 h after exposure to novelty stress (Fig. 1A).wo-way factorial ANOVA revealed significant effects oftress [F(1,28) = 37.90, p < 0.001], age [F(1,28) = 20.39,

< 0.001], and stress × age interaction [F(1,28) = 26.40, < 0.001] in male mice. Corticosterone secretion remainnchanged in aged female mice exposed to novelty stress,lthough the basal corticosterone levels in aged femaleice tended to be higher than those in young female mice

Fig. 1B). Further, two-way factorial ANOVA revealed a sig-ificant effect of age [F(1,27) = 20.70, p < 0.001], althoughhe effects of stress [F(1,27) = 1.85, n.s.], and stress × agenteraction [F(1,27) = 0.07, n.s.] were not significant inemale mice.

Food intake at 1 h after exposure to novelty stress onlylightly decreased in the young novelty stress-exposed malend female mice compared with young control mice (Fig. 1Cnd D). In contrast, food intake in the aged male micexposed to novelty stress remarkably decreased comparedith that in aged control male mice (Fig. 1C). Two-way

actorial ANOVA revealed significant effects of stress [1 h:(1,16) = 100.76, p < 0.001; 6 h: F(1,16) = 18.35, p < 0.001;4 h: F(1,15) = 16.95, p < 0.001] and age [1 h: F(1,16) = 13.15,

< 0.01; 6 h: F(1,16) = 24.63, p < 0.001; 24 h: F(1,15) = 24.72, < 0.001] at all-time points and stress × age interaction1 h: F(1,16) = 1.11, n.s.; 6 h: F(1,16) = 20.28, p < 0.001; 24 h:(1,15) = 4.76, p < 0.05] at 6 and 24 h in male mice. On thether hand, food intake in aged and stressed female micenly temporarily decreased and then returned to controlevels (Fig. 1D). Two-way factorial ANOVA revealed signif-cant effects of stress [1 h: F(1,16) = 143.20, p < 0.001; 6 h:(1,16) = 4.98, p < 0.05; 24 h: F(1,16) = 5.36, p < 0.05] at all-ime points, age [1 h: F(1,16) = 0.04, n.s.; 6 h: F(1,16) = 7.35,

< 0.05; 24 h: F(1,16) = 17.57, p < 0.001] at 6 and 24 h, andtress × age interaction [1 h: F(1,16) = 30.60, p < 0.001; 6 h:(1,16) = 1.33, n.s.; 24 h: F(1,16) = 1.06, n.s.] at 1 h in femaleice.

.2. Effects of CP-809101 on cumulative foodntake in young and aged mice

C administration of CP-809101 at a dose of 3.0 mg/kg tooung male (Fig. 2A) and female mice (Fig. 2B) significantlynhibited the cumulative food intake 1 h after treatment.

[sFb

C. Yamada et al.

n young female mice, CP-809101 (1.0 and 3.0 mg/kg) alsonhibited the 3-h cumulative food intake. Treatment of agedale mice with CP-809101 (1.0 and 3.0 mg/kg) suppressed 3-

food intake (Fig. 2C). Furthermore, the 1-h food intake inged male mice was also inhibited by administration of CP-09101 at a dose of 0.1 mg/kg. In contrast, compared withged control female mice, food intake in aged female micenly tended to decrease with CP-809101 treatment (Fig. 2D).

.3. Effects of 5-HT2CR antagonists on theecrease in cumulative food intake in aged micexposed to novelty stress

ompared with distilled water-treated aged male micexposed to stress throughout the experimental period, theumulative food intake in mice treated with SB2420846 mg/kg) or RKT (1000 mg/kg) was significantly increasedFig. 2E). In contrast, the cumulative food intake in agedemale mice was unaltered on administration of either ofhe two antagonists (Fig. 2F).

.4. Changes in the number of 5-HT2CR+ and-Fos+ cells in the PVN and amygdala of micexposed to novelty stress

ompared with aged control male mice, increased totalumbers of 5-HT2CR-positive cells in the PVN and 5-HT2CR-r c-Fos-positive cells in the amygdala were observed inged male mice exposed to novelty stress (Fig. 3A and B).he numbers of those cells in either the PVN or amygdala

n young male mice exposed to novelty stress remainednchanged. Two-way factorial ANOVA of the number of 5-T2CR-positive cells in both the PVN and amygdala revealedignificant effects of stress [PVN: F(1,9) = 10.55, p < 0.05;mygdala: F(1,9) = 5.56, p < 0.05] and stress × age interac-ion [PVN: F(1,9) = 5.53, p < 0.05; amygdala: F(1,9) = 5.76,

< 0.05], although the effect of age [PVN: F(1,9) = 4.06,.s.; amygdala: F(1,9) = 1.19, n.s.] was not significant.n addition, two-way factorial ANOVA of the number of-Fos-positive cells revealed significant effects of stressF(1,9) = 12.60; p < 0.01], age [F(1,9) = 13.36; p < 0.01], andtress × age interaction [F(1,9) = 7.22; p < 0.05] in themygdala, although the effects of stress [F(1,9) = 2.14;.s.], age [F(1,9) = 0.28; n.s.] and stress × age interactionF(1,9) = 5.09; n.s.] were not significant in the PVN. How-ver, in female mice, the number of these cells did nothange after stress exposure. Two-way factorial ANOVA ofhe number of 5-HT2CR-positive cells revealed a signifi-ant effect of age [F(1,9) = 7.02, p < 0.05], no significantffect of stress [F(1,9) = 0.09, n.s.] or stress × age inter-ction [F(1,9) = 0.67, n.s.] in the PVN, and no significantffect of stress [F(1,9) = 0.36, n.s.], age [F(1,9) = 0.21,.s.] or stress × age interaction [F(1,9) = 1.02, n.s.] in themygdala. Two-way factorial ANOVA of the number of c-os-positive cells revealed no significant effect of stressPVN: F(1,9) = 0.47, n.s.; amygdala: F(1,9) = 0.04, n.s.], age

PVN: F(1,9) = 0.15, n.s.; amygdala: F(1,9) = 2.71, n.s.], ortress × age interaction [PVN: F(1,9) = 0.15, n.s.; amygdala:(1,9) = 3.30, n.s.] in either the PVN or amygdala. The num-er of 5-HT2CR- and c-Fos co-positive cells significantly

Serotonin 2C receptor contributes to gender differences in stress-induced hypophagia in aged mice 85

Fig. 1 Plasma corticosterone levels and cumulative food intake after exposure to novelty stress. (A and B) Plasma corticosterone and

= 4—

osa

levels in male and female mice at 6 h after stress exposure. (C24 h after stress exposure. Data are presented as mean ± SEM (n

increased compared with aged control male mice in thePVN of aged male mice exposed to novelty stress (Figs.3A and 4B). However, expression of 5-HT2CR- and c-Fos co-

positive cells in the PVN of young male mice exposed tonovelty stress and in all female mice remain unchanged(Figs. 3A and 4B). Two-way factorial ANOVA of the number

mn[

D) Cumulative food intake in male and female mice 1, 6, and8). **, *** p < 0.01, 0.001 vs. age-matched control group.

f co-positive cells in the PVN revealed significant effects oftress [F(1,9) = 12.69, p < 0.01], age [F(1,9) = 7.02, p < 0.05],nd stress × age interaction [F(1,9) = 19.31, p < 0.01] in male

ice and no significant effect of stress [F(1,9) = 0.81,

.s.], age [F(1,9) = 1.58, n.s.], or stress × age interactionF(1,9) = 0.06, n.s.] in female mice. Further, compared with

86 C. Yamada et al.

Fig. 2 Effects of 5-HT2CR agonist or antagonists on cumulative food intake. (A—D) Cumulative food intake in young and aged maleand female mice after the administration of the 5-HT2CR agonist CP-809101 (0.1—3.0 mg/kg, SC). (E and F) Effects of the 5-HT2CRantagonists SB242084 (6 mg/kg, PO) or rikkunshito (RKT, 1000 mg/kg, PO) on cumulative food intake in aged male and female miceexposed to novelty stress. Data are presented as mean ± SEM (n = 5). #, ##, ### p < 0.05, 0.01, 0.001 vs. control group. *, *** p < 0.05,0

acmmi4c[sm

ttsAdp

.001 vs. stress group.

ged control male mice, the number of 5-HT2CR- and c-Foso-positive cells increased in the amygdala of aged maleice exposed to novelty stress (Figs. 3B and 4C). In youngale mice, exposure to novelty stress did not cause an

ncrease in the number of cells in the amygdala (Figs. 3B andC). Two-way factorial ANOVA of the number of co-positive

ells in the amygdala revealed significant effects of stressF(1,9) = 17.68, p < 0.01], age [F(1,9) = 25.17, p < 0.001] andtress × age interaction [F(1,9) = 12.49, p < 0.01] in maleice. Moreover, compared with aged control female mice,

aonf

he number of 5-HT2CR- and c-Fos co-positive cells inhe amygdala of aged female mice exposed to noveltytress increased (Figs. 3B and 4C). Two-way factorialNOVA of the number of co-positive cells in the amyg-ala revealed significant effects of stress [F(1,9) = 11.27,

< 0.01], age [F(1,9) = 5.55, p < 0.05], and stress × age inter-

ction [F(1,9) = 9.25, p < 0.05] in female mice. The numberf 5-HT2CR+ and c-Fos− cells or c-Fos+ and 5-HT2CR− cells didot change among all groups. There was only a trace of stainor these cells in the ARC (data not shown).

Serotonin 2C receptor contributes to gender differences in stress-induced hypophagia in aged mice 87

Fig. 3 Immunohistochemical analyses of 5-HT2CR and c-Fos expression in the hypothalamic paraventricular nucleus and amygdala.(A and B) Total number of 5-HT2CR+ cells (top panels), c-Fos+ cells (middle panels), and 5-HT2CR+ plus c-Fos+ (co-positive) cells(bottom panels) in the paraventricular nucleus (PVN) and amygdala of young and aged male and female mice at 6 h after stressexposure. White columns represent 5-HT2CR+ alone or c-Fos+ cells alone. Black columns represent 5-HT2CR+ plus c-Fos+ (co-positive)

, 0.0

ntasSlus

3aa

T

cells. Data are presented as mean ± SEM (n = 3—4). *, ** p < 0.05

3.5. Effects of aging and novelty stress exposureon plasma ghrelin levels, and effects of exogenousacylated ghrelin on the decrease in food intake

Although plasma acylated ghrelin levels in aged male miceexposed to novelty stress were significantly lower at 3 hand 6 h after exposure than in aged control mice, lev-els did not change in aged female mice (Fig. 5A and B).The decreased cumulative food intake in aged male miceat 1—3 h after exposure to novelty stress was completelyinhibited on IP administration of 10 nmol/kg exogenous acyl-ated ghrelin (Fig. 5C). Two-way factorial ANOVA revealedno significant effect of stress [F(1,14) = 3.51, n.s.], treat-ment [F(1,14) = 3.73, n.s.], or stress × treatment interaction[F(1,14) = 3.20, n.s.] at 1 h; a significant effect of treat-

ment [F(1,14) = 6.33, p < 0.05]; no significant effect ofstress [F(1,14) = 2.71, n.s.] or stress × treatment interac-tion [F(1,14) = 4.02, n.s.] at 3 h; no significant effect ofstress [F(1,14) = 2.45, n.s.] or treatment [F(1,14) = 3.35,

emCa

1 vs. age-matched control group.

.s.]; and a significant effect of stress × treatment interac-ion [F(1,14) = 4.75, p < 0.05] at 6 h. The decrease in plasmacylated ghrelin levels in aged male mice exposed to noveltytress was blocked by administering the 5-HT2CR antagonistsB242084 and RKT (Fig. 5D). The plasma acylated ghrelinevels in aged female mice exposed to novelty stress werenaffected by the administration of SB242084 (data nothown).

.6. Changes in hypothalamic ER�, ER�, andromatase mRNA expression levels in young andged mice

o clarify gender differences in food intake in aged mice

xposed to novelty stress, we focused on hypothala-ic ER�, ER�, and aromatase mRNA expression levels.ompared with young male mice, hypothalamic ER� andromatase mRNA expression levels significantly increased

8

iecmd(

3a

Fstn

8

n aged male mice, whereas ER� mRNA expression lev-ls remained unaltered in aged mice. In contrast, in

omparison with young female mice, ER� and aromataseRNA expression levels in aged female mice tended toecrease, but ER� mRNA expression levels were not alteredTable 1).

i

Nt

ig. 4 Images of 5-HT2CR and c-Fos expression in the hypothalamihowing immunoreactivity for c-Fos (left), 5-HT2CR (center), and bohe hypothalamic paraventricular nucleus (PVN) and amygdala of yovelty stress.

C. Yamada et al.

.7. Effects of aromatase inhibitor and ER�gonist on the decrease in cumulative food intake

n aged mice exposed to novelty stress

ext, we examined the role of aromatase and ER� onhe decreased food intake in aged male mice exposed to

c paraventricular nucleus and amygdala. (A) Examples of cellsth proteins (right). (B and C) 5-HT2CR and c-Fos expression in

oung and aged male and female mice at 6 h after exposure to

Serotonin 2C receptor contributes to gender differences in stress-induced hypophagia in aged mice 89

Fig. 5 Effects of ghrelin and 5-HT2CR antagonists on the decrease in cumulative food intake or plasma ghrelin levels. (A and B)Plasma acylated ghrelin levels at 3 and 6 h after stress exposure in aged male and female mice. (C) Effects of exogenous acylatedghrelin (10 nmol/kg, IP) on cumulative food intake in aged male mice. (D) Effects of 5-HT2CR antagonists [SB242084; 6 mg/kg, PO,rikkunshito (RKT); 1000 mg/kg, PO] on plasma acylated ghrelin levels in aged male mice at 3 h after stress exposure. (E) Effectsof an aromatase inhibitor (letrozole; 2.5 mg/kg, PO) and estrogen receptor (ER) � (DPN; 8 mg/kg, SC) on the decrease in the 6-hcumulative food intake in aged male mice exposed to novelty stress. (F and G) Cumulative food intake in young and aged maleand female mice after the administration of ER� agonist (PPT; 0.63, 1.25 mg/kg, SC). (H and I) Effects of SB242084 (6 mg/kg,PO) on decrease in food intake at 6 h after the administration of PPT (1.25 mg/kg, SC) in aged male and female mice. Data arepresented as mean ± SEM (n = 4—8). #, ##, ### p < 0.05, 0.01, 0.001 vs. control group. *, **, *** p < 0.05, 0.01, 0.001 vs. stress group,CP-809101-treated group, or PPT-treated group. DW; distilled water.

90

Table 1 Hypothalamic mRNA expression in young and agedmale and female mice.

Young Aged

MaleEstrogen receptor � 1.00 ± 0.07 1.43 ± 0.12##

Estrogen receptor � 1.00 ± 0.08 1.23 ± 0.08Aromatase 1.00 ± 0.22 1.91 ± 0.31#

FemaleEstrogen receptor � 1.00 ± 0.06 0.81 ± 0.13Estrogen receptor � 1.00 ± 0.05 1.07 ± 0.18Aromatase 1.00 ± 0.12 0.87 ± 0.18

noeazaSeo

3cc

AdctmfttiatppPa1eeaooId(dasaam

fsnm

4

Cnl2aaieat

hcbhrcs(rrmeetrftp2tatts

t2rapaeimltgiR

n = 4—8.#, ## p < 0.05, 0.01 vs. young mice.

ovelty stress. As observed in Fig. 5E, oral administrationf letrozole to aged male mice significantly inhibited nov-lty stress-induced decrease in food intake by 81% at 6 hfter the exposure. In contrast, in aged female mice, letro-ole administration did not significantly increase food intakefter novelty stress exposure compared with control mice.C administration of the selective ER� agonist DPN had noffects on novelty stress-induced hypophagia in aged maler female mice.

.8. Effects of ER� agonist alone ando-administered with 5-HT2CR antagonist onumulative food intake in aged mice

dministration of selective ER� agonist PPT at 1.25 mg/kgid not decrease food intake in young male and female miceompared with that in control mice (Fig. 5F and G). Fur-hermore, in aged male mice, treatment with the same oruch lower doses of PPT (0.63 and 1.25 mg/kg) suppressed

ood intake by 28—76% at 6 h and 24 h after administra-ion. Two-way factorial ANOVA revealed significant effects ofreatment [F(2,19) = 19.68, p < 0.001] and treatment × agenteraction [F(2,19) = 5.17, p < 0.05]; no significant effect ofge [F(1,19) = 0.04, n.s.] at 6 h; and significant effects ofreatment [F(2,19) = 18.11, p < 0.001], age [F(1,19) = 11.77,

< 0.01], and treatment × age interaction [F(2,19) = 3.87, < 0.05] at 24 h in male mice. However, administration ofPT barely decreased food intake in aged female mice,lthough it slightly decreased 6-h food intake at a dose of.25 mg/kg. Two-way factorial ANOVA revealed a significantffect of treatment [F(2,19) = 5.94, p < 0.01]; no significantffect of age [F(1,19) = 0.42, n.s.] or treatment × age inter-ction [F(2,19) = 0.76, n.s.] at 6 h; and no significant effectf treatment [F(2,19) = 1.99, n.s.], age [F(1,19) = 0.00, n.s.],r treatment × age interaction [F(2,19) = 0.15, n.s.] at 24 h.n young mice, SB242084 treatment did not suppress theecrease in food intake induced by PPT administrationFig. 5H and I). Administration of SB242084 prevented theecrease in the 6-h cumulative food intake in PPT-treatedged male mice (Fig. 5H). Two-way factorial ANOVA revealed

ignificant effects of treatment [F(2,34) = 22.10, p < 0.001]nd treatment × age interaction [F(2,34) = 7.05, p < 0.01]nd no significant effect of age [F(1,34) = 0.20, n.s.] in maleice. However, SB242084 did not affect food intake in aged

ttee

C. Yamada et al.

emale mice (Fig. 5I). Two-way factorial ANOVA revealed aignificant effect of treatment [F(2,34) = 9.59, p < 0.001] ando significant effect of age [F(1,34) = 2.05, n.s.] or treat-ent × age interaction [F(2,34) = 1.90, n.s.] in female mice.

. Discussion

ompared with young mice, in aged male mice, exposure toovelty stress caused a sustained increase in stress hormoneevels and a decrease in food intake that persisted until4 h after the exposure. Conversely, in aged female mice,lthough there was a significant decrease in food intake 1 hfter the stress exposure, it rapidly returned to a level sim-lar to that of the control group. These results indicate thatither aged female mice are more resistant to stress thanged male mice or that aged male mice are more vulnerableo stress.

In the present study, aged male mice exhibited markedlyigher blood levels of the stress marker corticosterone thanontrol mice even 6 h after stress loading. In contrast, theasal corticosterone levels of control female mice wereigher than those of control male mice, and their corticoste-one levels were not affected by stress loading. This result isonsistent with the findings of previous reports that demon-trated higher corticosterone levels in females than in malesGoel and Bale, 2010; Trainor et al., 2011). In addition, thisesult reproduced the findings of our previous study, whicheported a more persistent stress response in aged maleice compared with the response in young mice (Nahata

t al., 2013). However, the basal blood corticosterone lev-ls of young control female mice in our study may be higherhan the levels reported in other recent studies. Corticoste-one levels may vary depending on the circadian rhythm,asting state, and measurement method. The effects ofhese factors cannot be ruled out in the present study orrevious studies (Trainor et al., 2013; Van Ginhoven et al.,010). In addition, the mice used as the control group inhe present study were individually housed for 1 week forcclimatization, and they were fasted for 18 h; these fac-ors may have served as additional stressors that contributedo the differences in corticosterone levels between thetudies.

Previous studies have shown that 5-HT2CR activation leadso anxious behavior and suppression of food intake (Gatch,003; Halford et al., 2007). Our current study revealed thatesponsiveness to a selective 5-HT2CR agonist in young mice,s indicated by decreased food intake, tended to be moreronounced in females than in males. This finding is in goodgreement with a previous observation by Iwasaki-Sekinot al. (2009). It is likely that 5-HT2CR responsiveness is highn young female rodents. In contrast, administration in agedale mice led to a marked decrease in food intake at the

owest dose of 0.1 mg/kg. However, in aged female mice,here was no decrease in food intake. These results sug-est that aging leads to functional upregulation of 5-HT2CRn male but not in female mice. Furthermore, SB242084 andKT clearly inhibited the decrease in food intake, although

hese drugs had no effect on food intake when administeredo aged female mice. Based on these results, we hypoth-size that in aged male mice, exposure to novelty stressnhances the binding of 5-HT to 5-HT2CR, thereby markedly

tres

peatewdwfiiacoso

eiastfycedmtats

ia2e2ttpsyesoaeErmtimotw

Serotonin 2C receptor contributes to gender differences in s

decreasing food intake. We believe that the function of 5-HT2CR can decrease with age in female mice, leading to no5-HT2CR-dependent inhibition of food intake.

Intracerebral 5-HT is released from the terminals ofneurons that project from the raphe nucleus to the stress-related areas of the brain. Several lines of evidence showthat aging leads to a decrease in the basal levels or biosyn-thesis of 5-HT in the hypothalamus (De la Fuente et al., 2003)or the amygdala (Miura et al., 2002). However, biosynthesisof intracerebral 5-HT was enhanced in rats exposed to socialisolation, resulting in increased 5-HT concentrations (Miuraet al., 2002). This finding suggests that exposure to stress inaged male mice can lead to excessive release of 5-HT.

To confirm the upregulation of 5-HT2CR signaling in agedmale mice during exposure to novelty stress, we per-formed immunohistochemical analysis for 5-HT2CR and c-Fosprotein in the PVN, amygdala, and ARC. A change in c-Fos expression in stress-related cerebral regions is widelyused as a stress indicator as well as plasma corticoste-rone levels (Zangenehpour and Chaudhuri, 2002). In thepresent study, the number of c-Fos-positive cells at 6 hafter stress loading in young mice was similar to that inyoung control mice. Although acute c-Fos expression fol-lowing stress exposure was not confirmed in this study,c-Fos expression was previously reported to shortly increaseafter novelty stress exposure (Handa et al., 1993). Further-more, increased plasma corticosterone levels were foundonly 30 min after stress loading in young mice (Saegusaet al., 2011). In the present study, plasma corticosteronelevels at 6 h after stress loading in young mice were alsothe same as those in young control mice. That is, at thetime point of 6 h after stress loading, c-Fos expression inyoung mice may have already recovered to the same levelas in the control group. The co-positive cell counts for 5-HT2CR and c-Fos in the PVN and amygdala did not increasewith age in male mice. This is consistent with the findingsthat show that aging itself does not stimulate activationof 5-HT neurons but that 5-HT remains at normal or lowerthan normal levels (De la Fuente et al., 2003). Inciden-tally, co-positive cell counts for 5-HT2CR and c-Fos proteinin the PVN and amygdala markedly increased in aged malemice after exposure to novelty stress, although there wasno significant change in positive cell counts for 5-HT2CR orc-Fos alone in the same brain area. Exposing aged malemice to novelty stress increases the expression of 5-HT2CRmRNA in the PVN (Nahata et al., 2013), suggesting thatin aged male mice, exposure to novelty stress promotes5-HT2CR protein synthesis in stress-specific neurons, acti-vating neurons that express 5-HT2CR. In contrast, there wasno change in 5-HT2CR and c-Fos co-positive cell counts inthe PVN of aged female mice exposed to stress, parallelingthe results for food intake and corticosterone release. The5-HT2CR and c-Fos co-positive cell count in the amygdalaof aged female mice exposed to novelty stress was signifi-cantly higher than that of aged control mice, as was the casefor aged male mice. These results suggest that aged femalemice also exhibit sustained negative emotional responsesbesides decreased feeding behavior. In addition, the fact

that we found only trace expression of 5-HT2CR and c-Fos inthe ARC at 6 h after the stress exposure (data not shown)suggests that continuous activation of 5-HT2CR in the ARC isunlikely.

wire

s-induced hypophagia in aged mice 91

Ghrelin is the only peripheral orexigenic hormone thatlays a central role in physiological feeding behavior (Kojimat al., 1999). However, whether ghrelin levels fluctuate inged or female mice has not been investigated. We foundhat plasma levels of acylated ghrelin in aged male micexposed to novelty stress significantly decreased, but thereas no change in the levels in aged female mice. Whenecreased peripheral acylated ghrelin was supplementedith exogenous ghrelin in aged male mice, the decreased

ood intake returned to normal levels, indicating that stress-nduced ghrelin insufficiency may contribute to the decreasen food intake. Of particular interest, the decrease in plasmacylated ghrelin levels observed in aged male mice wasompletely reversed by administration of the 5-HT2CR antag-nists. These results indicate that the enhanced 5-HT2CRignal may decrease food intake by inhibiting the secretionf acylated ghrelin.

In this study, we used 6-week- and 79-week-old mice. Thestrous cycle of C57BL/6 mice was reported to be 4 daysn young mature mice and just less than once a month inged mice. The estrous cycle of 6-week-old mice used in thistudy may be constant or partly in an irregular phase prioro the stabilization (Nelson et al., 1982). Sex-dependent dif-erences in peripheral estradiol levels were not observed inoung or aged mice, although we did not measure the estrousycle of each animal and the values might be only for ref-rence (Supplemental materials). We were unable to obtainirect evidence that could conclusively attribute the abnor-al stress response observed in aged mice to aging itself or

o an indirect effect through peripheral sex hormone imbal-nce caused by aging. Therefore, further adequate attentiono the effect of estrous cyclicity may be required for thetudy of feeding behavior in female mice in the future.

Many previous studies have demonstrated that estradiolnhibits food intake via ER�s and inhibits HPA axis activitynd anxiety via ER�s (Asarian and Geary, 2013; Handa et al.,012). Moreover, these effects are proportional to brainstrogen levels and not to blood estrogen levels (Liu et al.,011). Estrogen produced from androgen by aromatase inhe brain of male mice plays an important role in stress reac-ions and the control of food intake (Liu et al., 2011). In theresent study, the expression of aromatase and ER� mRNAignificantly increased in aged male mice compared withoung mice. However, no significant variations in ER� mRNAxpression were observed. In aged female mice, expres-ion of aromatase and ER� mRNA did not increase. Basedn these results, we assumed that decreased food intakend HPA axis activation in aged male mice were caused byxposure to novelty stress through an increased level ofR� expression and excessive production of estrogen as aesult of increased aromatase in the brain. From the above-entioned results, we first evaluated the effect of letrozole

o clarify the involvement of aromatase in decreased foodntake due to stress. Furthermore, to eliminate the involve-ent of ER�, we examined the effect of DPN administration

n food intake. Although estrogen levels in the brain orhe periphery after administration of an aromatase inhibitorere not measured in the present study, our hypothesis

as strongly supported by the fact that decreased food

ntake in aged male mice exposed to novelty stress wasesolved with the aromatase inhibitor letrozole and that thisffect was not observed in female mice. In addition, the

9

ai

lPcmypaaepontpttt

o5d

R

Tol

C

HYtNa

A

TfGfcIi(

A

Sfj

R

A

A

B

D

d

D

D

D

D

D

E

G

G

H

H

H

H

H

H

2

dministration of the ER� agonist DPN did not affect foodntake in male mice.

To determine the effect of ER� on food intake in stress-oaded aged animals, we administered the ER� agonistPT to mice. Enhanced functional activity of ER�, indi-ated by decreased food intake, was demonstrated in agedice, particularly in aged male mice compared with that in

oung mice. Estradiol has been reported to enhance 5-HT2CRrotein synthesis in the dorsal raphe region (Hendersonnd Bethea, 2008), caudal brainstem (Rivera et al., 2012),nd hypothalamus (Santollo et al., 2012). The anorexigenicffect of low doses of PPT observed in aged male mice in theresent study significantly decreased by the 5-HT2CR antag-nist, suggesting that the increased ER� signals observed inovelty stress-loaded aged male mice decreased food intakehrough an increase in 5-HT2CRs. In addition, consideringrevious studies stating that estrogen increases 5-HT2CRs inhe brain, the increase in the PVN of aged male mice seemso have occurred as a result of excessive ER� activation dueo increased estrogen and ER�s in the brain.

In conclusion, we believe that ER� activation thatccurred in aged male mice exposed to novelty stress caused-HT2CR hypersensitivity, leading to HPA axis activation andecreased food intake.

ole of the funding source

here was no impact from funding source(s) on any aspectf the work with the current manuscript (design, data col-ection, analysis, interpretation, writing, or submission).

onflict of interest statement

. Takeda received grant support from Tsumura & Co.; C.amada, C. Sadakane, M. Nahata, Y. Saegusa, and T. Hat-ori are employed by Tsumura & Co.; and K. Nakagawa,. Okubo, and S. Ohnishi have no conflicts to declare. Theuthors have no competing interests.

cknowledgments

his work was supported in part by the Research Fundingor Longevity Sciences (25—3) from the National Center foreriatrics and Gerontology (NCGG), Japan, a Grant-in-Aid

or research (22590676) from the Japanese Ministry of Edu-ation, Culture, Sports, Science and Technology, a grant fornterdisciplinary Project for Psychosomatological Researchn Hokkaido University, and a grant from Tsumura & Co.Ibaraki, Japan).

ppendix A. Supplementary data

upplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.psyneuen.2015.02.006.

eferences

rndt, S.S., Laarakker, M.C., van Lith, H.A., van der Staay, F.J.,Gieling, E., Salomons, A.R., van’t Klooster, J., Ohl, F., 2009.

C. Yamada et al.

Individual housing of mice—–impact on behaviour and stressresponses. Physiol. Behav. 97, 385—393.

sarian, L., Geary, N., 2013. Sex differences in the physiologyof eating. Am. J. Physiol. Regul. Integr. Comp. Physiol. 305,R1215—R1267.

echtholt, A.J., Hill, T.E., Lucki, I., 2007. Anxiolytic effect ofserotonin depletion in the novelty-induced hypophagia test. Psy-chopharmacology 190, 531—540.

e la Fuente, M., Hernanz, A., Medina, S., Guayerbas, N., Fernan-dez, B., Viveros, M.P., 2003. Characterization of monoaminergicsystems in brain regions of prematurely ageing mice. Neu-rochem. Int. 43, 165—172.

e Mello Cruz, A.P., Pinheiro, G., Alves, S.H., Ferreira, G., Mendes,M., Faria, L., Macedo, C.E., Motta, V., Landeira-Fernandez, J.,2005. Behavioral effects of systemically administered MK-212are prevented by ritanserin microinfusion into the basolateralamygdala of rats exposed to the elevated plus-maze. Psy-chopharmacology 182, 345—354.

iefenbach, G.J., Goethe, J., 2006. Clinical interventions for late-life anxious depression. Clin. Interv. Aging 1, 41—50.

iniz, B.S., Reynolds 3rd, C.F., Butters, M.A., Dew, M.A., Firmo,J.O., Lima-Costa, M.F., Castro-Costa, E., 2014. The effect ofgender, age, and symptom severity in late-life depression on therisk of all-cause mortality: the Bambui cohort study of aging.Depress. Anxiety 31, 787—795.

ombrovski, A.Y., Szanto, K., Duberstein, P., Conner, K.R., Houck,P.R., Conwell, Y., 2008. Sex differences in correlates of sui-cide attempt lethality in late life. Am. J. Geriatr. Psychiatry16, 905—913.

ryden, S., Wang, Q., Frankish, H.M., Williams, G., 1996. Differen-tial effects of the 5-HT 1B/2C receptor agonist mCPP and the5-HT1A agonist flesinoxan on hypothalamic neuropeptide Y inthe rat: evidence that NPY may mediate serotonin’s effects onfood intake. Peptides 17, 943—949.

ulawa, S.C., Hen, R., 2005. Recent advances in animal models ofchronic antidepressant effects: the novelty-induced hypophagiatest. Neurosci. Biobehav. Rev. 29, 771—783.

hlert, U., Gaab, J., Heinrichs, M., 2001. Psychoneuroendocrinolog-ical contributions to the etiology of depression, posttraumaticstress disorder, and stress-related bodily disorders: the roleof the hypothalamus—pituitary—adrenal axis. Biol. Psychol. 57,141—152.

atch, M.B., 2003. Discriminative stimulus effects of m-chlorophenylpiperazine as a model of the role of serotoninreceptors in anxiety. Life Sci. 73, 1347—1367.

oel, N., Bale, T.L., 2010. Sex differences in the serotonergicinfluence on the hypothalamic—pituitary—adrenal stress axis.Endocrinology 151, 1784—1794.

alford, J.C., Harrold, J.A., Boyland, E.J., Lawton, C.L., Blundell,J.E., 2007. Serotonergic drugs: effects on appetite expressionand use for the treatment of obesity. Drugs 67, 27—55.

anda, R.J., Mani, S.K., Uht, R.M., 2012. Estrogen receptors andthe regulation of neural stress responses. Neuroendocrinology96, 111—118.

anda, R.J., Nunley, K.M., Bollnow, M.R., 1993. Induction of c-fos mRNA in the brain and anterior pituitary gland by a novelenvironment. NeuroReport 4, 1079—1082.

anda, R.J., Weiser, M.J., 2014. Gonadal steroid hormones and thehypothalamo—pituitary—adrenal axis. Front. Neuroendocrinol.35, 197—220.

ayashi, A., Suzuki, M., Sasamata, M., Miyata, K., 2005. Agonistdiversity in 5-HT(2C) receptor-mediated weight control in rats.Psychopharmacology 178, 241—249.

eisler, L.K., Cowley, M.A., Tecott, L.H., Fan, W., Low, M.J., Smart,J.L., Rubinstein, M., Tatro, J.B., Marcus, J.N., Holstege, H.,

Lee, C.E., Cone, R.D., Elmquist, J.K., 2002. Activation of cen-tral melanocortin pathways by fenfluramine. Science (New York,N.Y.) 297, 609—611.

tres

P

R

R

S

S

S

S

S

S

T

T

T

V

Z

Serotonin 2C receptor contributes to gender differences in s

Heisler, L.K., Zhou, L., Bajwa, P., Hsu, J., Tecott, L.H., 2007. Sero-tonin 5-HT(2C) receptors regulate anxiety-like behavior. GenesBrain Behav. 6, 491—496.

Henderson, J.A., Bethea, C.L., 2008. Differential effects of ovariansteroids and raloxifene on serotonin 1A and 2C receptor proteinexpression in macaques. Endocrine 33, 285—293.

Hotta, M., Shibasaki, T., Arai, K., Demura, H., 1999. Corticotropin-releasing factor receptor type 1 mediates emotional stress-induced inhibition of food intake and behavioral changes in rats.Brain Res. 823, 221—225.

Iwasaki-Sekino, A., Mano-Otagiri, A., Ohata, H., Yamauchi, N.,Shibasaki, T., 2009. Gender differences in corticotropin and cor-ticosterone secretion and corticotropin-releasing factor mRNAexpression in the paraventricular nucleus of the hypothalamusand the central nucleus of the amygdala in response to footshockstress or psychological stress in rats. Psychoneuroendocrinology34, 226—237.

Kennett, G.A., Wood, M.D., Bright, F., Trail, B., Riley, G., Holland,V., Avenell, K.Y., Stean, T., Upton, N., Bromidge, S., Forbes, I.T.,Brown, A.M., Middlemiss, D.N., Blackburn, T.P., 1997. SB 242084,a selective and brain penetrant 5-HT2C receptor antagonist.Neuropharmacology 36, 609—620.

Kojima, M., Hosoda, H., Date, Y., Nakazato, M., Matsuo, H., Kan-gawa, K., 1999. Ghrelin is a growth-hormone-releasing acylatedpeptide from stomach. Nature 402, 656—660.

Kornstein, S.G., 1997. Gender differences in depression: implica-tions for treatment. J. Clin. Psychiatry 58 (Suppl. 15), 12—18.

Liu, J., Hu, P., Qi, X.R., Meng, F.T., Kalsbeek, A., Zhou, J.N.,2011. Acute restraint stress increases intrahypothalamic oestra-diol concentrations in conjunction with increased hypothalamicoestrogen receptor beta and aromatase mRNA expression infemale rats. J. Neuroendocrinol. 23, 435—443.

Meng, F.T., Ni, R.J., Zhang, Z., Zhao, J., Liu, Y.J., Zhou, J.N.,2011. Inhibition of oestrogen biosynthesis induces mild anxietyin C57BL/6J ovariectomized female mice. Neurosci. Bull. 27,241—250.

Miura, H., Qiao, H., Ohta, T., 2002. Influence of aging and socialisolation on changes in brain monoamine turnover and biosyn-thesis of rats elicited by novelty stress. Synapse (New York, NY)46, 116—124.

Morley, J.E., 1993. Nutrition and the older female: a review. J. Am.Coll. Nutr. 12, 337—343.

Morley, J.E., 2010. Nutrition and the aging male. Clin. Geriatr. Med.26, 287—299.

Nahata, M., Muto, S., Nakagawa, K., Ohnishi, S., Sadakane, C.,Saegusa, Y., Iizuka, S., Hattori, T., Asaka, M., Takeda, H.,2013. Serotonin 2C receptor antagonism ameliorates novelty-induced hypophagia in aged mice. Psychoneuroendocrinology38, 2051—2064.

Nelson, J.F., Felicio, L.S., Randall, P.K., Sims, C., Finch, C.E., 1982.A longitudinal study of estrous cyclicity in aging C57BL/6J mice:

I Cycle frequency, length and vaginal cytology. Biol. Reprod. 27,327—339.

Nonogaki, K., 2008. Ghrelin and feedback systems. Vitam. Horm.77, 149—170.

Z

s-induced hypophagia in aged mice 93

aykel, E.S., 1994. Life events, social support and depression. ActaPsychiatr. Scand. Suppl. 377, 50—58.

ivera, H.M., Santollo, J., Nikonova, L.V., Eckel, L.A., 2012. Estra-diol increases the anorexia associated with increased 5-HT(2C)receptor activation in ovariectomized rats. Physiol. Behav. 105,188—194.

owe, J.W., Kahn, R.L., 1987. Human aging: usual and successful.Science (New York, NY) 237, 143—149.

aegusa, Y., Takeda, H., Muto, S., Nakagawa, K., Ohnishi, S.,Sadakane, C., Nahata, M., Hattori, T., Asaka, M., 2011.Decreased plasma ghrelin contributes to anorexia following nov-elty stress. Am. J. Physiol. Endocrinol. Metab. 301, E685—E696.

antollo, J., Wiley, M.D., Eckel, L.A., 2007. Acute activation ofER alpha decreases food intake, meal size, and body weightin ovariectomized rats. Am. J. Physiol. Regul. Integr. Comp.Physiol. 293, R2194—R2201.

antollo, J., Yao, D., Neal-Perry, G., Etgen, A.M., 2012. Middle-aged female rats retain sensitivity to the anorexigenic effect ofexogenous estradiol. Behav. Brain Res. 232, 159—164.

eeman, T.E., Robbins, R.J., 1994. Aging and hypothalamic—pituitary—adrenal response to challenge in humans. Endocr. Rev.15, 233—260.

hin, J.A., Yang, S.J., Jeong, S.I., Park, H.J., Choi, Y.H., Park, E.M.,2013. Activation of estrogen receptor beta reduces blood—brainbarrier breakdown following ischemic injury. Neuroscience 235,165—173.

teenbergen, H.L., Heinsbroek, R.P., Van Hest, A., Van de Poll, N.E.,1990. Sex-dependent effects of inescapable shock administra-tion on shuttlebox-escape performance and elevated plus-mazebehavior. Physiol. Behav. 48, 571—576.

akeda, H., Sadakane, C., Hattori, T., Katsurada, T., Ohkawara, T.,Nagai, K., Asaka, M., 2008. Rikkunshito, an herbal medicine,suppresses cisplatin-induced anorexia in rats via 5-HT2 receptorantagonism. Gastroenterology 134, 2004—2013.

rainor, B.C., Pride, M.C., Villalon Landeros, R., Knoblauch, N.W.,Takahashi, E.Y., Silva, A.L., Crean, K.K., 2011. Sex differences insocial interaction behavior following social defeat stress in themonogamous California mouse (Peromyscus californicus). PLoSONE 6, e17405.

rainor, B.C., Takahashi, E.Y., Campi, K.L., Florez, S.A., Green-berg, G.D., Laman-Maharg, A., Laredo, S.A., Orr, V.N., Silva,A.L., Steinman, M.Q., 2013. Sex differences in stress-inducedsocial withdrawal: independence from adult gonadal hormonesand inhibition of female phenotype by corncob bedding. Horm.Behav. 63, 543—550.

an Ginhoven, T.M., Van Den Berg, J.W., Dik, W.A., Ijzermans, J.N.,De Bruin, R.W., 2010. Preoperative fasting induces protectionagainst renal ischemia/reperfusion injury by a corticosterone-independent mechanism. Transpl. Int. 23, 1171—1178.

angenehpour, S., Chaudhuri, A., 2002. Differential induction anddecay curves of c-fos and zif268 revealed through dual activity

maps. Brain Res. Mol. Brain Res. 109, 221—225.

orrilla, E.P., Tache, Y., Koob, G.F., 2003. Nibbling at CRF receptorcontrol of feeding and gastrocolonic motility. Trends Pharmacol.Sci. 24, 421—427.