Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1

http://dx.doi.org/100006-8993/& 2014 El

Abbreviations: AM

the calcium calmod

glucocorticoid recep

II-Cre (see MethodMR, mineralocorti

thalamus; vHPC, vnCorresponding autE-mail address: j

Research Report

Forebrain glucocorticoid receptor gene deletionattenuates behavioral changes and antidepressantresponsiveness during chronic stress

Lauren Jacobsonn

Center for Neuropharmacology and Neuroscience, Albany Medical College, Mail Code 146, Albany, NY 12208, USA

a r t i c l e i n f o

Article history:

Accepted 30 July 2014

Stress is an important risk factor for mood disorders. Stress also stimulates the secretion of

glucocorticoids, which have been found to influence mood. To determine the role of

Available online 26 August 2014

Keywords:

Glucocorticoid

Depression

Anxiety

Antidepressant

Defeat

Stress

.1016/j.brainres.2014.07.05sevier B.V. All rights res

, morning; BLA, baso

ulin kinase II promotertor deletion derived on a

s); GR, glucocorticoid re

coid receptor(s); NAc, n

entral hippocampus; ilPFhor. Fax: 1 518 262 5799acobsL@mail.amc.edu

a b s t r a c t

forebrain glucocorticoid receptors (GR) in behavioral responses to chronic stress, the

present experiments compared behavioral effects of repeated social defeat in mice with

forebrain GR deletion and in floxed GR littermate controls. Repeated defeat produced

alterations in forced swim and tail suspension immobility in floxed GR mice that did not

occur in mice with forebrain GR deletion. Defeat-induced changes in immobility in floxed

GR mice were prevented by chronic antidepressant treatment, indicating that these

behaviors were dysphoria-related. In contrast, although mice with forebrain GR deletion

exhibited antidepressant-induced decreases in tail suspension immobility in the absence

of stress, this response did not occur in mice with forebrain GR deletion after defeat. There

were no marked differences in plasma corticosterone between genotypes, suggesting that

behavioral differences depended on forebrain GR rather than on abnormal glucocorticoid

secretion. Defeat-induced gene expression of the neuronal activity marker c-fos in the

ventral hippocampus, paraventricular thalamus and lateral septum correlated with

genotype-related differences in behavioral effects of defeat, whereas c-fos induction in

the nucleus accumbens and central and basolateral amygdala correlated with genotype-

related differences in behavioral responses to antidepressant treatment. The dependence

of both negative (dysphoria-related) and positive (antidepressant-induced) behaviors on

forebrain GR is consistent with the contradictory effects of glucocorticoids on mood, and

implicates these or other forebrain regions in these effects.

& 2014 Elsevier B.V. All rights reserved.

4erved.

lateral amygdale; CamKII-Cre, Transgene expressing Cre recombinase under control of; CeA, central amygdale; Cort, corticosterone; FBGRKO-T29-1, Mice with forebrain

pure C57BL/6 strain from a commercial founder transgenic for calcium calmodulin kinase

ceptor(s); HPA, hypothalamicpituitaryadrenocortical; LS, lateral septum;

ucleus accumbens; PVN, paraventricular hypothalamus; pvThal, paraventricular

C, infralimbic prefrontal cortex.

http://dx.doi.org/10.1016/j.brainres.2014.07.054http://dx.doi.org/10.1016/j.brainres.2014.07.054http://dx.doi.org/10.1016/j.brainres.2014.07.054http://dx.doi.org/10.1016/j.brainres.2014.07.054http://crossmark.crossref.org/dialog/?doi=10.1016/j.brainres.2014.07.054&domain=pdfmailto:jacobsL@mail.amc.eduhttp://dx.doi.org/10.1016/j.brainres.2014.07.054

Table 1 Results of the social interaction test in floxed GRand FBGRKO-T29-1mice on day 15 of control cage expo-sures (Control) or repeated social defeat (Defeat) fromExperiment 1. Data are time (sec) mice spent interactingwith a novel CD-1 male mouse that was confined in aperforated plastic box.

Floxed GR FBGRKO-T29-1

Control Defeat Control Defeat

59715 N8 2474n N11 66719 N8 3577 N11

n Po0.05 vs. control in the same genotype.

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1110

1. Introduction

A variety of studies have found that stress increases the risk ofclinically significant emotional disorders (Caspi et al., 2003;Jacobs et al., 1990; Maes et al., 2000). The mechanisms under-lying the psychiatric risks of stress are largely undefined,although they have been linked, for example, to interactionwith genetic factors such as serotonin transporter isoforms(Caspi et al., 2003). However, genetic make-up is difficult tomodify, indicating a need to explore other mechanisms bywhich stress can influence mental health.

Stress is also a major stimulus for glucocorticoid secretion bythe hypothalamicpituitaryadrenocortical (HPA) axis. Glucocor-ticoids have been found to have a variety of negative effects onmood ranging from anxiety and depression to psychosis, even inindividuals lacking a family or personal history of psychiatricdisease. Such effects have been correlated with elevated gluco-corticoid levels in Cushing's syndrome and in patients receivingglucocorticoids for immunologic disorders (Brown, 2009; Kennaet al., 2011). Glucocorticoids are also often elevated in depression,a finding that has been attributed to deficits in glucocorticoidreceptors responsible for HPA feedback inhibition (Ising et al.,2007). Although the significance of the increase in glucocorticoidsremains a matter of debate, there is limited but intriguingevidence to suggest that glucocorticoids can contribute todepression symptoms (Rakofsky et al., 2009; Wolkowitz et al.,2009). Consistent with the clinical literature, glucocorticoids havealso been shown to have depression- and anxiety-like effects inrodents (Mitchell and Meaney, 1991; Sterner and Kalynchuk,2010; Tronche et al., 1999; Veldhuis et al., 1985; Wei et al., 2004).However, glucocorticoids have also been found to have mood-elevating effects (Brown 2009; Kenna et al., 2011), and depressionhas also been suggested to be due to deficits in glucocorticoidsreceptors mediating positive mood (Sudheimer et al., 2013).Glucocorticoids are therefore a logical and compelling connec-tion between stress and affective disease risk.

Glucocorticoids can bind in brain to either the higher-affinity mineralocorticoid receptor (MR) or the lower affinityglucocorticoid receptor (GR). GR is widely expressed through-out the brain and thought to mediate the effects of elevatedlevels of glucocorticoids, whereas MR is concentrated in butnot limited to limbic structures (Jacobson, 2005) and thought tobe more sensitive to low glucocorticoid levels (Jacobson, 2005).Although there is some evidence that MR can affect emotion-related behavior in rodents (Lai et al., 2007; Mostalac-Preciadoet al., 2011; Rozeboom et al., 2007; Smythe et al., 1997; Wuet al., 2013), depression- and anxiety-like symptoms are mostfrequently evoked in humans and animals at elevated levels ofglucocorticoids (Brown, 2009; Kenna et al., 2011; Schellinget al., 2006), which are more likely to activate GR. Therefore,GR are the most plausible candidate to be involved in themood effects of glucocorticoids.

GR in a variety of forebrain regions have been implicatedin affective dysfunction (Boyle et al., 2005; Calfa et al., 2007;Kolber et al., 2008; Korte et al., 1996; McKlveen et al., 2013;Sudheimer et al., 2013; Yang et al., 2006). Mice with forebrainGR deletion have been created by transgenic expression ofCre recombinase under control of the calcium calmodulinkinase II (CamKII) promoter in floxed GR mice. The original

model of forebrain GR deletion, derived on a mixed-strainbackground from the T50 founder line of the CamKII-Cretransgene, was reported to have a depression-like phenotype,consisting of increased depression-like behavior and elevatedHPA activity relative to that in floxed GR mice (Boyle et al.,2005). However, this depression-like phenotype was notfound in mice with forebrain GR deletion derived on a pureC57BL/6 background from the T29-1 founder line of theCamKII-Cre transgene, even though forebrain GR deletionwas at least as extensive (Vincent et al., 2013). The lattermouse model, hereafter referred to as FBGRKO-T29-1 (Vincentet al., 2013), offers a convenient model to test the role offorebrain GR in the behavioral effects of chronic stress,without potential confounds from baseline differences inHPA activity or depression-like behavior. The current experi-ments compared the effects of repeated social defeat stress orcontrol cage exposures, with and without antidepressanttreatment, on depression-related behavior in FBGRKO-T29-1mice and their genetic controls, floxed GR mice. Althoughchronic stress elicited changes in behavior in floxed GR micethat were opposite to those conventionally interpreted ascorrelates of depression, these changes were associated withsocial aversion, a depression-like behavior (Krishnan et al.,2007), and were reversed by chronic antidepressant treat-ment, suggesting that behavioral alterations were dysphoria-related. FBGRKO-T29-1 mice failed to exhibit these stress-related changes in behavior and were resistant to the effectsof antidepressant treatment during chronic stress, suggestingGR involvement in both negative (dysphoria-related) andpositive (antidepressant-induced) effects on mood.

2. Results

2.1. Experiment 1: effects of forebrain GR deletion onbehavioral and HPA responses to repeated social defeat

Social interaction with a novel conspecific has been used as ameasure of depression-like behavior, with lower levels ofinteraction interpreted as greater depression-like behavior(Krishnan et al., 2007). Interaction with a novel mouse exhibiteda significant main effect of defeat (F1,348.852; P0.0054) but nosignificant main effects of genotype or genotype x defeatinteraction. Defeated floxed GR mice exhibited significantly lesstime investigating the novel mouse than did control floxed GRmice (Table 1). There was a similar trend for defeated FBGRKO-T29-1 mice to exhibit less interaction with a novel mouse, but

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 111

this effect was not significant at the post-hoc level (P0.097).There were no significant differences among genotypes ordefeat groups in the amount of time mice spent interactingwith an empty box as a neutral target (data not shown).

Sucrose preference, a measure of pleasure-seeking behavior(Willner, 2005), was similar between floxed GR and FBGRKO-T29-1 mice after control cage exposures. However, sucrose prefer-ence exhibited a significant main effect of defeat (F1,3119.758;P0.0001), with defeated floxed GR but not defeated FBGRKO-T29-1 mice exhibiting significantly higher sucrose preferencerelative to their corresponding genotype controls (Fig. 1A). Therewere no significant main effects of genotype or genotypedefeat interaction on sucrose preference.

Immobility in the forced swim test exhibited a significantmain effect of defeat (F1,347.516; P0.0097) and a significant

ControlDefeat

SucrosePreference(%)

Immo-bility

(sec)

AMPlasmaCort( g/dl)

Floxed GR FBGRKO-T29-1

0

20

40

60

80

100 *

0

100

200

300

*

0

1

2

3

4

5

Fig. 1 Effects of repeated social defeat on (A) sucrosepreference, (B) forced swim immobility, and (C) basalmorning plasma corticosterone in floxed GR and FBGRKO-T29-1 mice, respectively measured on days 21, 23, and 30days of control cage exposures (Control; white bars) orrepeated social defeat (Defeat; gray bars) in Experiment 1.N8 (control FBGRKO-T29-1 and floxed GR mice) and 11(defeated FBGRKO-T29-1 and floxed GR mice) except forsucrose preference, in which some measurements were lostto spillage (N7 control FBGRKO-T29-1 and floxed GR mice,10 defeated floxed GR mice). nPo0.05 vs. control in the samegenotype.

defeatgenotype interaction (F1,345.246; P0.0283), with nosignificant main effect of genotype alone. Defeated floxed GRmice displayed significantly lower levels of immobility in theforced swim test than did control floxed GR mice (Fig. 1B).Control FBGRKO-T29-1 mice tended to be less immobile thanwere control floxed GR mice, but this difference was notsignificant (P40.1; Fig. 1B). However, in contrast to floxed GRmice, FBGRKO-T29-1 mice did not exhibit any further decreasesin immobility between control and defeated groups (Fig. 1B).

Basal circadian nadir plasma corticosterone did not exhi-bit any significant main effects of defeat, genotype, orgenotypedefeat interaction (Fig. 1C).

2.2. Experiment 2: effects of forebrain GR deletion onbehavioral and HPA responses to antidepressant treatmentduring repeated social defeat

As in Experiment 1, there was a significant main effect ofdefeat in Experiment 2 to reduce social interaction with anovel mouse (F1,275.360; P0.0284), but there were nosignificant effects of genotype, imipramine treatment or anyinteraction among defeat, genotype, and treatment on socialinteraction (Table 2). There were also no significant effects ofgenotype, defeat, or antidepressant treatment on interactionwith the box alone (not shown) or on locomotor activity inthe test arena (Table 3).

There were no significant main effects of genotype, defeat,or antidepressant treatment on sucrose preference aftersaline or imipramine treatment during repeated defeat orcontrol cage exposures in Experiment 2 (Table 4).

There was a significant genotype treatment interactionon tail suspension immobility in Experiment 2 (F1,274.835;P0.0366), although no other main effects or interactions weresignificant. Resembling the forced swim test results in Experi-ment 1, tail suspension immobility was reduced in floxed GRmice subjected to repeated defeat in Experiment 2 (Fig. 2).Although chronic imipramine treatment did not affect tailsuspension immobility of control floxed GR mice, imipraminetreatment of defeated floxed GRmice restored their immobilityto levels observed in saline-treated, control floxed GR mice(Fig. 2). Also similar to the results of Experiment 1, saline-treated FBGRKO-T29-1 mice did not exhibit any changes inimmobility as a result of repeated social defeat (Fig. 2). ControlFBGRKO-T29-1 mice did exhibit significant decreases in tailsuspension immobility after chronic imipramine treatment,

Table 2 Results of the social interaction test in floxed GRand FBGRKO-T29-1mice on day 25 of saline or imipra-mine treatment during control cage exposures (Control)or repeated social defeat (Defeat) in Experiment 2. Dataare time (s) mice spent interacting with a novel CD-1 malemouse that was confined in a perforated plastic box.

Floxed GR FBGRKO-T29-1

Saline Imipramine Saline Imipramine

Control 35724 1076 43720 1774N4 N3 N4 N4

Defeat 772 1074 1178 372N4 N6 N5 N5

Table 4 Sucrose preference on day 21 of saline orimipramine treatment during control cage exposures(Control) or repeated social defeat (Defeat) in Experiment2. Data indicate the percent of total fluid intake repre-sented by consumption of 1% sucrose.

Floxed GR FBGRKO-T29-1

Saline Imipramine Saline Imipramine

Control 5777 5174 6275 52710N4 N3 N4 N4

Defeat 6778 3979 46710 5479N4 N6 N5 N5

Immo-bility(sec)

Floxed GR FBGRKO-T29-1

Control, SalineDefeat, SalineControl, Imipramine

Defeat, Imipramine

4 3 4 6 4 4 5 5

Fig. 2 Tail suspension immobility in floxed GR andFBGRKO-T29-1 mice from Experiment 2 on day 29 of salineor imipramine treatment during control cage exposures(Control; white or hatched bars) or repeated social defeat(Defeat; gray or black bars). Group Ns are indicated by thenumbers within the data bars. nPo0.05 vs. control, saline inthe same genotype. #Po0.05 vs. defeat, imipramine in thesame genotype.

Table 5 Circadian nadir (morning; AM) plasma corti-costerone in floxed GR and FBGRKO-T29-1 mice inExperiment 2. Samples were collected by sub-mandibu-lar venipuncture within 1 h of lights-on on day 28 ofsaline or imipramine treatment during control cageexposures (Control) or repeated social defeat (Defeat).

AM plasma corticosterone (g/dl)

Floxed GR FBGRKO-T29-1

Saline Imipramine Saline Imipramine

Control 0.870.2 0.670.1 1.070.2 0.770.1N4 N3 N4 N4

Defeat 0.870.3 2.670.9 0.870.1 0.970.2N4 N6 N5 N5

Table 3 Locomotor activity in floxed GR and FBGRKO-T29-1 mice on day 25 of saline or imipramine during control cageexposures (Control) or repeated social defeat (Defeat) in Experiment 2. Data are the total distance traveled (cm) during a2.5 min test session in a 3543 cm basin.

Floxed GR FBGRKO-T29-1

Saline Imipramine Saline Imipramine

Control 3677137 4187158 5287118 470771N4 N3 N4 N4

Defeat 385773 421785 362743 372799N4 N6 N5 N5

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1112

unlike control floxed GR mice (Fig. 2). However, the lowerlevels of immobility observed after imipramine treatment incontrol FBGRKO-T29-1 mice did not occur in defeated FBGRKO-T29-1 mice treated with imipramine (Fig. 2).

Basal morning (circadian nadir) corticosterone was mea-sured 2 d before the end of the experiment (day 28 imipra-mine treatment), and stress-induced corticosterone wasmeasured in samples collected at death, 30 min after a social

defeat in all groups. Basal corticosterone was not collected onthe day of the final defeat in order to avoid stress effects fromblood sampling on subsequent HPA responses to defeat.There were no significant effects of genotype, defeat, anti-depressant treatment, or interaction among these factors onbasal plasma corticosterone in Experiment 2 (Table 5).Although basal morning corticosterone appeared to be ele-vated in defeated floxed GR mice, this difference was notsignificant. There were also no significant main effects orinteractions of genotype, defeat, or treatment on stressplasma corticosterone sampled 30 min after an acute socialdefeat in all mice (Table 6).

To localize GR-expressing forebrain regions potentiallyaccounting for genotypic differences in behavioral responsesto defeat or imipramine treatment, gene expression of theneuronal activity marker c-fos was analyzed by in situ hybri-dization in brains collected after subjecting both control andpreviously defeated mice in Experiment 2 to an acute defeat.Significant differences between control and repeatedlydefeated mice that were present in one genotype but notthe other were interpreted to indicate brain regions poten-tially involved in genotype-related differences in behavioralresponses to defeat (saline-treated groups) or antidepressanttreatment (imipramine-treated groups). In all brain regionsstudied, there was a significant main effect of repeated defeatto reduce levels of c-fos expression induced by a final defeat(Table 7 and Fig. 3). Other ANOVA results are detailed belowand in Table 7.

In the infralimbic prefrontal cortex, where GR deletion hasbeen shown to increase depression-like behavior (McKlveenet al., 2013), c-fos induction was significantly lower in

Table 6 Stress-induced plasma corticosterone evoked by an acute social defeat in all floxed GR and FBGRKO-T29-1 mice inExperiment 2. Mice had been treated with saline or imipramine during the prior 29 days of control cage exposures (Control)or repeated social defeat (Defeat). Samples were collected by decapitation 30 min after the defeat. No injections wereadministered or basal corticosterone samples collected prior to the defeat sample in order to avoid confounding effects onsubsequent responses to defeat.

30' Post-defeat plasma corticosterone (g/dl)

Floxed GR FBGRKO-T29-1

Saline Imipramine Saline Imipramine

Control 26.173.4 28.471.0 32.074.5 26.075.4N4 N3 N4 N4

Defeat 23.174.3 21.873.7 25.674.9 16.874.3N4 N6 N5 N5

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 113

saline-treated defeated mice vs. saline-treated controls in bothgenotypes (Fig. 3, ilPFC). Imipramine did not significantly alterc-fos induction in the infralimbic prefrontal cortex in controlor defeated mice of either genotype (Fig. 3, ilPFC). There wereno significant differences among any group in c-fos mRNAlevels in the prelimbic prefrontal cortex (not shown).

In the shell of the nucleus accumbens, an area involved inmotivated behavior and social interaction (Barik et al., 2013;Nestler and Carlezon, 2006), there was little difference in c-fosgene expression among groups of floxed GR mice (Fig. 3, NAc).However, c-fos induction in FBGRKO-T29-1 mice was signifi-cantly lower in imipramine-treated defeated mice comparedto imipramine-treated control mice (Fig. 3, NAc). There wasalso a significant genotypedefeat treatment interaction(Table 7) such that c-fos induction in the nucleus accumbensshell was marginally lower (P0.075) in imipramine-treated,defeated FBGRKO-T29-1 mice than in imipramine-treated,defeated floxed GR mice (Fig. 3, NAc).

In the lateral septum, which has been implicated as aglucocorticoid target in mediating defeat-induced anxiety(Calfa et al., 2007), the pattern of c-fos induction appeared tobe similar between genotypes. However, significant differencesoccurred only in floxed GR mice, between saline-treatedcontrol mice and saline-treated defeated mice (Fig. 3, LS).

The paraventricular hypothalamus was analyzed becauseit expresses most of the releasing factors responsible forhypothalamicpituitaryadrenocortical axis activity (Jacobson,2005). In the paraventricular hypothalamus, c-fos inductiontended, although not significantly, to be higher in FBGRKO-T29-1 vs. floxed GR saline-treated control mice and was associatedwith a significant post-hoc difference between saline-treatedcontrol and saline-treated defeated FBGRKO-T29-1 mice (Fig. 3,PVN). However, despite significant main effects of genotype andgenotypedefeat treatment interaction (Table 7), there wereno significant differences in PVN c-fos expression between floxedGR and FBGRKO-T29-1 mice at the post-hoc level.

The posterior paraventricular thalamus, which is involvedin habituation of responses to chronic stress (Bhatnagar et al.,2002, 2003), did not exhibit significant differences in c-fosinduction among groups of floxed GR mice (Fig. 3, pvThal).However, FBGRKO-T29-1 mice treated with saline exhibitedsignificantly lower levels of c-fos expression in defeated vs.control mice; c-fos expression in FBGRKO-T29-1 mice was notaffected by imipramine treatment (Fig. 3, pvThal).

The central and basolateral amygdala were investigatedbecause of their role in fear, anxiety, and adaptation ofresponses to chronic stresses such as defeat (Johansen et al.,2011; Kolber et al., 2008; Markham et al., 2010; Nikulina et al.,2004). In the central amygdala, both genotypes exhibitedsignificantly lower levels of c-fos induction in saline-treateddefeated mice vs. saline-treated controls (Fig. 3, CeA). Imipra-mine treatment did not increase c-fos expression in defeatedmice of either genotype. However, in FBGRKO-T29-1 mice,central amygdala c-fos expression was significantly lower inimipramine-treated defeated mice compared to that inimipramine-treated control mice (Fig. 3, CeA). In the basolat-eral amygdala, c-fos expression was also significantly lower inimipramine-treated control vs. imipramine-treated defeatedFBGRKO-T29-1 mice (Fig. 3, BLA).

The CA1 field of the ventral hippocampus was analyzedbecause GR are prominently expressed in CA1 of floxed GRbut not FBGRKO-T29-1 mice (Boyle et al., 2005, 2006; Vincentet al., 2013), and because the ventral hippocampus has limbicconnections that are both distinct from the connections ofthe dorsal hippocampus and relevant to emotion-relatedbehavior (Fanselow and Dong 2010; Felix-Ortiz et al., 2013;Markham et al., 2010). Induction of c-fos gene expression inCA1 ventral hippocampus did not differ significantly amonggroups of floxed GR mice, although c-fos tended to be lower(P0.054) in defeated vs. control floxed GR mice treated withimipramine (Fig. 3, vHPC). In contrast, the ventral CA1hippocampus of FBGRKO-T29-1 mice exhibited significantlylower c-fos expression in saline-treated defeated vs. controlmice but had similar levels of c-fos expression betweenimipramine-treated defeated and control mice (Fig. 3, vHPC).

3. Discussion

These experiments demonstrate that forebrain GR deletioncan have state-specific effects on behavior. Forebrain GRdeletion, at least in the FBGRKO-T29-1 mice used here(Vincent et al., 2013), did not affect baseline behavior andattenuated behaviors, particularly the immobility response toinescapable situations, induced by the chronic stress ofrepeated defeat. Furthermore, although the current experi-ments confirm prior findings (Boyle et al., 2005; Vincent et al.,2013) that forebrain GR deletion does not impair, and may

Table 7 ANOVA results for c-fos gene expression data from Experiment 2. Residual degrees of freedom differ for someregions because damaged or missing sections occasionally precluded obtaining measurements from a given mouse. Pvalues less than 0.05 are bolded for clarity.

Region Factor Fdf P

Infralimbic prefrontal cortex Genotype 0.6841,25 0.4161Defeat 28.2301,25 o0.0001Treatment 0.6611,25 0.4238GenotypeDefeat 0.0221,25 0.8844GenotypeTreatment 0.00021,25 0.9892DefeatTreatment 7.7551,25 0.0101GenotypeDefeatTreatment 2.6791,25 0.1142

Nucleus accumbens shell Genotype 1.2271,27 0.2779Defeat 8.2831,27 0.0077Treatment 0.5511,27 0.4644GenotypeDefeat 5.6081,27 0.0253GenotypeTreatment 0.1011,27 0.7535DefeatTreatment 0.1491,27 0.7023GenotypeDefeatTreatment 5.4291,27 0.0275

Lateral septum Genotype 0.8531,27 0.3639Defeat 16.6891,27 0.0004Treatment 2.1301,27 0.1560GenotypeDefeat 0.0121,27 0.9133GenotypeTreatment 0.0811,27 0.7779DefeatTreatment 0.4841,27 0.4925GenotypeDefeatTreatment 0.0071,27 0.9358

Paraventricular hypothalamus Genotype 6.1081,26 0.0203Defeat 32.0851,26 o0.0001Treatment 6.3541,26 0.0182GenotypeDefeat 3.0151,26 0.0943GenotypeTreatment 8.4891,26 0.0073DefeatTreatment 2.5861,26 0.1199GenotypeDefeatTreatment 5.3611,26 0.0288

Paraventricular thalamus Genotype 0.6031,27 0.4442Defeat 16.3601,27 0.0004Treatment 0.9401,27 0.3408GenotypeDefeat 0.4371,27 0.5141GenotypeTreatment 0.2441,27 0.6250DefeatTreatment 4.8011,27 0.0373GenotypeDefeatTreatment 0.1481,27 0.7035

Central amygdala Genotype 0.0031,27 0.9577Defeat 44.9111,27 o0.0001Treatment 6.1981,27 0.0192GenotypeDefeat 0.0031,27 0.9592GenotypeTreatment 0.2481,27 0.6224DefeatTreatment 3.9551,27 0.0569GenotypeDefeatTreatment 0.4691,27 0.4991

Basolateral amygdala Genotype 0.00041,27 0.9836Defeat 12.5271,27 0.0015Treatment 0.0421,27 0.8399GenotypeDefeat 0.0021,27 0.9633GenotypeTreatment 0.8901,27 0.3540DefeatTreatment 0.9991,27 0.3265GenotypeDefeatTreatment 1.9841,27 0.1704

Ventral hippocampus Genotype 0.0261,27 0.8738Defeat 7.9911,27 0.0087Treatment 0.0361,27 0.8518GenotypeDefeat 0.4661,27 0.5008GenotypeTreatment 1.2391,27 0.2755DefeatTreatment 2.9691,27 0.0963GenotypeDefeatTreatment 0.5231,27 0.4758

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1114

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 115

even increase, behavioral sensitivity to antidepressantsunder baseline conditions, forebrain GR loss preventedantidepressant-induced changes in immobility behavior dur-ing chronic stress. The behavioral effects of forebrain GRdeletion during chronic stress occurred in the absence ofmarked differences in glucocorticoid secretion betweenFBGRKO-T29-1 and floxed GR mice, suggesting that altera-tions in behavior were more likely to depend on differencesin GR expression than on differences in HPA activity. Differ-ential activation of GR-expressing forebrain regions afterdefeat, as assessed by c-fos gene expression, suggested thatthe ventral hippocampus, paraventricular thalamus andlateral septum might be involved in the behavioral resilience

Floxed GR FBGRKO-T29-1

Control, SalineDefeat, SalineControl, ImipramineDefeat, Imipramine

LS *

BLA

CeA**

NAc

pvThal * *

vHPC *

ilPFC** **

PVN*

of FBGRKO-T29-1 mice to repeated social defeat, whereas thenucleus accumbens shell and central and basolateral amyg-dala might be involved in the resistance of FBGRKO-T29-1mice to antidepressant treatment during chronic defeat stress.

Forebrain GR deletion was associated with fewer changesin behavior after chronic social defeat, even if these changesdid not resemble conventional definitions of depression-likebehavior (discussed below). Floxed GR but not FBGRKO-T29-1mice exhibited increased activity in the forced swim and tailsuspension tests after repeated defeat. Increased forced swimand tail suspension activity was not due to general hyperlo-comotion, since overall activity was not increased. Althoughsucrose preference only exhibited significant effects of defeatin Experiment 1, even this behavioral change was presentonly in floxed GR and not in FBGRKO-T29-1 mice.

Antidepressant sensitivity was also influenced by forebrainGR. Imipramine had the expected action of an antidepressantto decrease immobility in the tail suspension test, but only incontrol FBGRKO-T29-1 mice. Although prior evidence of acuteimipramine effects on immobility in unstressed floxed GRmice (Vincent et al., 2013) appears to conflict with the lack ofresponse to chronic imipramine treatment in control floxed GRmice, peak antidepressant levels after acute injection are likelyto be higher than those at the time of testing in the presentexperiments, since testing occurred at least 16 h after theprevious day's injection. The greater sensitivity of unstressedFBGRKO-T29-1 mice to imipramine is consistent with theoriginal report that mice with forebrain GR deletion respondwith changes in immobility to imipramine doses that do notaffect floxed GR mice (Boyle et al., 2005). The sensitivity toimipramine in FBGRKO-T29-1 mice was abrogated by repeatedsocial defeat, since imipramine-treated, defeated FBGRKO-T29-1 mice exhibited neither the decrease in immobilityinduced by imipramine in FBGRKO-T29-1 controls nor theincrease in immobility elicited by imipramine in defeatedfloxed GR mice. Thus, immobility behavior in FBGRKO-T29-1mice was responsive to antidepressants in unstressed but notchronically stressed conditions. These findings indicate thatforebrain glucocorticoid receptors are involved not only in

Fig. 3 Results of in situ hybridization analysis of c-fos geneexpression in brains collected from floxed GR and FBGRKO-T29-1 mice 30min after a social defeat in Experiment 2.Groups are the same as in Fig. 2; mice had been treated withsaline or imipramine for the last 29 d of a 60 d period of dailycontrol cage exposures (Control) or social defeat (Defeat).Panels respectively depict semi-quantitative gray levelreadings taken from (top to bottom) the infralimbic prefrontalcortex (ilPFC), nucleus accumbens shell (NAc), lateral septum(LS), paraventricular hypothalamus (PVN), paraventricularthalamus (pvThal), central amygdala (CeA), basolateralamygdala (BLA), and ventral hippocampus (vHPC).Representative phosphorimager autoradiograms are shownto the right of each graph, with white outlines indicating theregion analyzed. Ns are the same as in Fig. 3 except for thePVN (N4 for FBGRKO-T29-1 Control, Saline) and infralimbicPFC (N4 for FBGRKO-T29-1 Defeat, Imipramine). nPo0.05 vs.control, saline in the same genotype. Po0.05 vs. control,imipramine in the same genotype.

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1116

changes in immobility behavior evoked by chronic stress, butalso in the response of immobility behavior to antidepressantsduring chronic stress. The lack of imipramine effects on socialinteraction or sucrose preference in the current experimentsmay have been due to administering antidepressant duringrather than after defeat, as has been done in other studies(Berton et al., 2006; Blugeot et al., 2011; Razzoli et al., 2011).

There were no detectable differences in basal or stress-induced glucocorticoids between genotypes. Although thestrong paraventricular hypothalamus c-fos induction indefeated FBGRKO-T29-1 mice suggested that greater HPAactivity might be occurring, PVN c-fos induction is not apredictor of HPA activity (Ginsberg et al., 2003) and did notdiffer between genotypes. While it cannot be excluded thatmore or different sampling points might have revealeddivergent HPA activity between FBGRKO-T29-1 and floxedGR mice, prior measurements of adrenocorticotropin as wellas corticosterone in FBGRKO-T29-1 mice have been consis-tently similar to those of floxed GR mice under a variety ofconditions (Vincent et al., 2013). Furthermore, forebrain GRhave not been found to influence changes in HPA activityinduced by chronic stress (Furay et al., 2008), and preliminarydata in FBGRKO-T29-1 mice corroborate these findings (Jacob-son, unpublished findings). Thus, it is most likely that fore-brain GR loss, rather than differences in HPA activity orhabituation, accounted for the behavioral differencesbetween floxed GR and FBGRKO-T29-1 mice after defeat.

Mapping c-fos gene expression after exposing all mice to asocial defeat suggested areas in which GR loss might modifybehavioral responses to defeat stress or antidepressant treat-ment. Genotype-associated differences between saline-treatedcontrol and defeated mice, potentially indicating mechanismsfor the lack of stress-induced changes in immobility behaviorin FBGRKO-T29-1 mice, were found in the ventral hippocam-pus, the paraventricular thalamus, and the lateral septum. Theparaventricular thalamus is involved in suppression of beha-vioral and HPA responses to chronic stress (Bhatnagar et al.,2002, 2003), while the septum and ventral hippocampus areinvolved in anxiety-like behavior, an effect influenced byseptal GR (Calfa et al., 2007; Markham et al., 2010). Thus,differential activity in these areas would be in accord with therelative lack of stress-induced changes in immobility behaviorin FBGRKO-T29-1 mice. However, only the ventral hippocam-pus displayed differences in the overall pattern of c-fosinduction between FBGRKO-T29-1 and floxed GR mice, withonly FBGRKO-T29-1 mice exhibiting significant differencesbetween saline-treated control and defeated groups. There-fore, ventral hippocampal activity may be most closely relatedto the resilience of FBGRKO-T29-1 mice to stress-inducedabnormalities in immobility responses to inescapable situa-tions. This interpretation is in keeping with evidence that theventral hippocampus can regulate fear- and anxiety-relatedbehavior (Calfa et al., 2007; Felix-Ortiz et al., 2013; Markhamet al., 2010; Sotres-Bayon et al., 2012).

Genotype-related differences in defeat-induced c-fos expres-sion between imipramine-treated control and imipramine-treated defeated mice, potentially indicating pathways under-lying the antidepressant resistance of stress-induced immobilitybehavior in defeated FBGRKO-T29-1 mice, occurred in thenucleus accumbens shell, central amygdala, and basolateral

amygdala. These regions have all been found to be activatedby defeat (Markham et al., 2010; Nikulina et al., 2004). The centraland basolateral amygdala, as well as central amygdala GR, areinvolved in fear responses (Johansen et al., 2011; Kolber et al.,2008). The more pronounced differences in basolateral andcentral amygdala c-fos expression between saline- andimipramine-treated FBGRKO-T29-1 mice after defeat could beconnected to the impairment of imipramine responsiveness byfear-related experiences of defeat. However, the pattern of c-fosexpression differed most strongly between genotypes in thenucleus accumbens shell. Gene expression of c-fos in thenucleus accumbens shell was essentially similar in all floxedGR groups but was lower in imipramine-treated FBGRKO-T29-1defeated mice compared not only to imipramine-treatedFBGRKO-T29-1 control mice, but also, marginally, to imipra-mine-treated, floxed GR defeated mice. Thus, the nucleusaccumbens shell seems to be the best candidate of these threeregions to account for the loss of imipramine effects on immo-bility in FBGRKO-T29-1 mice during chronic defeat stress. Invol-vement of the nucleus accumbens in GR-related effects of defeatis consistent with findings that dopamine release in the nucleusaccumbens correlates with GR-dependent changes in defeat-induced behavior (Barik et al., 2013).

Limitations of this study include small group sizes inExperiment 2, phenotypic discrepancies between FBGRKO-T29-1 mice and the originally reported model of forebrain GRdeletion, limited sampling times, and atypical effects of stresson behavior in floxed GR mice. Larger group sizes might havepermitted confirmation in Experiment 2 of behavioral effectsthat were significant in Experiment 1, such as changes insucrose preference, and might also have allowed more con-clusive or extensive identification of brain regions exhibitingchanges in c-fos expression related to stress or antidepressantresponse. Nevertheless, when behavioral alterations were sig-nificant in floxed GR mice, such changes were reliably missingin FBGRKO-T29-1 mice, and 6 of the 9 brain regions examinedshowed genotype-related differences in c-fos induction.

In these and previous experiments (Vincent et al., 2013),FBGRKO-T29-1 mice did not exhibit the higher baseline immo-bility, lower sucrose preference, or elevated HPA activity origin-ally reported for mixed-strain mice with forebrain GR deletionderived from the T50 CamKII-Cre founder line, but otherwisehave GR deletion that is similar to that of the original model(Boyle et al., 2005; Vincent et al., 2013). The reasons for thediscrepancies in baseline behavior and endocrine functionbetween the two lines are unclear, but may be related todifferences in strain, founder, central amygdala GR loss, ormineralocorticoid receptor expression (Vincent et al., 2013).Nevertheless, the current findings still indicate importantinfluences of forebrain GR on emotion-related immobilitybehavior during chronic stress that are independent of, or morerobust than, the effects of forebrain GR on baseline behavior.

The main effect of prior defeat to decrease defeat-inducedc-fos expression in all brain regions was consistent withhabituation of responses to repeated stress (Kollack-Walkeret al., 1999; Weinberg et al., 2010). Since only one time pointwas studied after defeat or antidepressant treatment, it ispossible that differences in c-fos or behavior between floxedGR and FBGRKO-T-29-1 mice are due to differential habitua-tion to chronic stress, rather than to differences in inherent

Experiment 1

Experiment 2

Daily social defeat/ control handling

15 21 23 30

SIT SP FST AMkill

Daily social defeat/ control handling

585551 59 60

AMCort

SITSP TST Kill 30post-defeat

Imipramine/ saline treatment

30

Fig. 4 Diagram of the design of Experiments 1 and 2 (topand bottom, respectively). Numbers above the bars indicatethe days on which tests indicated below the bars wereperformed. SIT, social interaction test; SP, sucrosepreference; FST, forced swim test; AM, morning; AM Cort,morning plasma corticosterone sample; TST, tailsuspension test.

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 117

responses. This possibility seems unlikely because of the longperiod during which mice were stressed and because of thesimilar patterns in immobility behavior between control anddefeated mice in Experiments 1 (30 d) and 2 (60 d), by whichtime habituation would be expected to be complete. Even ifresults are attributable to differences in habituation, thesefindings indicate an important impact of forebrain GR on thehabituation process.

In floxed GR mice, the decreased immobility and, whenobserved, increased sucrose preference after repeated defeatappeared to conflict with the increases in forced swim immo-bility and decreases in sucrose preference that are oftenreported after chronic stress (Blugeot et al., 2011; Krishnanet al., 2007; Willner, 2005). Although there is no ready explana-tion for the contradictory direction of these stress-inducedchanges, there is precedent in the literature for the paradoxicalbehavioral changes observed in the current experiments.Chronic stress, including repeated social defeat, has been foundby other investigators to decrease forced swim immobility orincrease sucrose preference (Bowens et al., 2012; Der-Avakianet al., 2014; Dubreucq et al., 2012; Platt and Stone, 1982; Razzoliet al., 2011; Swiergiel et al., 2007; Ver Hoeve et al., 2013; Willner,2005). These atypical behavioral responses to stress, althoughless common, have been proposed to be part of a continuum ofbehavioral repertoires that is represented to varying degrees inall study populations and may account for inconsistent resultsin any given behavioral assay (Willner, 2005). There is alsoevidence that lower immobility in the forced swim or tailsuspension test reflects an increase in anxiety-like behaviorrather than a decrease in depression-like behavior (Boyle et al.,2006; Van Dijken et al., 1992). The possibility that decreasedimmobility is a dysphoria-related behavior is supported by itsconsistent association with reductions in social interaction, anadditional marker of depression- or anxiety-like behavior(Berton et al., 2006; File, 1980; Krishnan et al., 2007), and byfindings, here and in other studies (Montkowski et al., 1995),that the decreased immobility is reversed by chronic antide-pressant treatment.

The present experiments implicate forebrain GR in boththe expression and the antidepressant reversal of chronicstress effects on dysphoria-related immobility. The depen-dence of both negative (stress-induced) and positive (anti-depressant-induced) behaviors on forebrain glucocorticoidreceptors is consistent with the evidence that glucocorticoidscan have mood-elevating as well as depressive effects(Brown, 2009; Kenna et al., 2011). Defining the brain regions,cell types, and downstream targets mediating these contra-dictory effects of glucocorticoids will aid in reducing the riskof affective dysfunction from major life stress or immuno-suppressive glucocorticoid therapy.

4. Experimental procedures

4.1. Animals

All animal use was approved by the Institutional Animal Careand Use Committee of Albany Medical College and was con-sistent with the standards of the National Institutes of HealthGuide for the Care and Use of Laboratory Animals (Institute of

Laboratory Animal Resources, Commission on Life Sciences,National Research Council, 2010) and European CommissionDirective 2010/63/EU. Mice were housed on a 12:12 light cycle(lights-on, 6:30 am) and had free access to food and water.Floxed GR mice were derived from C57BL/6 founders homo-zygous for a floxed exon 2 and were generously donated by Dr.Louis Muglia (University of Cincinnati; (Brewer et al., 2003)).FBGRKO-T29-1 mice were bred on a pure C57 background fromcrosses between female floxed GR mice and male floxed GRmice hemizygous for the CamKII-Cre transgene (Vincent et al.,2013). The latter were derived from commercially available,C57BL/6 CamKII-Cre transgenic mice (Jackson Laboratoriesstock number 005359, Bar Harbor, ME) from the T-29-1 founderline (Tsien et al., 1996). Only male FBGRKO-T29-1 and floxed GRmice littermates were used for experiments and were 1.52.5months old at the time of use.

4.2. Experiments

The relative timing of experimental manipulations and testsis diagramed in Fig. 4. In Experiment 1, mice were killed bydecapitation in the basal state within 3 h of lights-on after 30days of defeat. The 30 day period of defeat was used to allowtime to conduct tests of behaviors in addition to social defeat,changes in which were not observed until at least 21 days ofdefeat.

In Experiment 2, daily defeats were continued for another29 d while mice were injected ip with either 30 mg/kg imi-pramine or saline. Injections were given once per day aftereach defeat or control cage exposure experience and after anybehavioral testing or blood sampling was performed. Mice inExperiment 2 were killed by decapitation on day 60, 1 d aftertheir last injection and 30 min after an acute social defeat inall control and defeat groups. The 30 min time point waspreviously shown to be appropriate for analysis of both HPAhormones and c-fos gene expression (Bowens et al., 2012).

Social defeat was performed by placing a FBGRKO-T29-1 orfloxed GR intruder mouse into the home cage of a singly-housed, CD-1 retired male breeder (Charles River, Wilming-ton, MA). Intruder mice were initially left in the resident'scage for 5 min; however, to avoid injuries to the intruder,

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1118

encounters were subsequently limited to the time the CD-1resident first attacked and bit the intruder. Latency to the firstattack was typically 3060 s, and all attacks elicited consistentsubmissive or fleeing behavior from all intruder mice, regard-less of genotype. Only CD-1 males exhibiting attack latenciesless than 1 min were used as resident aggressors. Intruderswere returned to their home cage and housed individuallyafter each daily attack. Preliminary experiments in whichintruders were co-housed with CD-1 residents, separated by aclear, perforated partition, produced similar behavioraleffects (data not shown). Intruders, designated as the defeatgroups in the rest of text and figures, were rotated amongdifferent residents on a daily basis to minimize habituation ofany behavioral interaction. Control mice of each genotypewere placed for 30 s each day into a cage from which theresident male had been removed. Controls (designated as thecontrol groups in the rest of text and figures) were alsorotated among different cages each day.

Daily defeats and control cage exposures were alwaysperformed after any behavioral tests or blood sampling on thesame day. Defeats and control cage exposures occurred atvariable times in the light phase (Krishnan et al., 2007) between1 and 10 h after lights-on, as dictated by the need to completebehavioral or hormonal testing beforehand and share access tothe animal roomwith other users. However, defeats and controlcage exposures were performed during the same 1.5 h timewindow on any given day for all mice in the experiment.

4.3. Behavioral tests

Behavioral tests were typically performed within the first 3 hof the light phase under quiet conditions. Social interactiontests required somewhat longer but were completed with 5 hof lights-on. All testing chambers were cleaned with MB-10disinfectant (Quip Laboratories, Wilmington, DE) in betweenmice. Tests were performed on separate days in the order ofleast (sucrose preference or social interaction) to most stress-ful (forced swim or tail suspension), as diagramed in Fig. 4.Tests in Experiment 2 were timed to occur after at least 3weeks of imipramine treatment; other differences in thesequence of tests between Experiments 1 and 2 resulted fromthe need to fit experimental procedures around caretakerschedules in the animal room.

4.3.1. Social interaction testSocial interaction was measured on day 15 of social defeat inExperiment 1 and on day 25 of imipramine treatment (55 ddefeat) in Experiment 2 as a positive control for the effects ofdefeat to induce social aversion, a measure of depression-likebehavior (Berton et al., 2006; Krishnan et al., 2007). Mice wereplaced in a 3543 cm basin with a clear, perforated box(6.5812 cm3) in the opposite corner. For the first 2.5 minof the trial, the box was empty, providing a neutral interactiontarget. During the second 2.5 min, an identical perforated boxcontaining an unfamiliar CD-1 male mouse as a novel socialtarget was placed in the corner, allowing mice to haveolfactory, auditory, and visual contact without physical inter-action. The time mice spent in the 10 cm interaction zonesurrounding the box, without and with the unfamiliar mouse,was recorded with Ethovision 8.0 software (Noldus, Leesburg,

VA). The total distance traveled in the trial with the empty boxwas also recorded as measure of overall locomotor activity.

4.3.2. Sucrose preferenceSucrose preference was tested on day 21 of social defeat inExperiment 1 and on day 21 of imipramine treatment (51 ddefeat) in Experiment 2. Mice were given access to two 50-mltubes containing either water or 1% sucrose for 24 h beforeand 24 h after the specified day, with the position of the tubesswitched to avoid position preference effects. Mice were notfood-deprived during sucrose preference measurements.Sucrose preference was expressed as the percentage ofsucrose vs. total fluid consumed.

4.3.3. Forced swim testImmobility was scored on day 23 of social defeat in Experi-ment 1 in a 5-min forced swim that had been preceded by a15-min pre-swim the day before. Mice were placed in a 1-literbeaker in 2571 1C water, and swimming behavior wasrecorded with Ethovision 8.0. Immobility was scored afterblinding the genotype and treatment of the mice. The 2-dayforced swim test was originated by Porsolt et al. in rats andremains a standard procedure for measuring depression-likebehavior (Cryan et al., 2002; Porsolt et al., 1977b). Although thetest was shortened by Porsolt et al. to a single swim for micePorsolt et al. (1977a), both paradigms give equivalent results(Cryan et al., 2002). We chose to use the 2-day paradigm inmice because glucocorticoid levels at the time of the first swimhave been shown to increase immobility in the test swim 24 hlater in a GR-dependent manner (Veldhuis et al., 1985).

4.3.4. Tail suspension testA tail suspension test was performed in Experiment 2 on day29 of imipramine treatment (59 d defeat) as an additional testto confirm that genotype-related differences in immobilitywere consistent across different paradigms for measuringemotion-related activity (Cryan et al., 2002). Mice were tapedby their tail to a bar and allowed to hang upside-down for8 min, during which time struggling behavior was recordedwith Ethovision 8.0. The time mice hung completely motion-less was scored from Ethovision videos by a trained observerblind to genotype and treatment groups. Manual scoring wasnecessary because the software did not distinguish activestruggling from the passive swinging that can occur from themomentum of previous activity.

4.4. Plasma corticosterone assay

All samples were collected within 45 s of touching the cage.In Experiment 1, morning (circadian nadir) plasma corticos-terone samples were collected by decapitation within 3 h oflights-on, after 30 d of social defeat or control cage exposure.In Experiment 2, plasma corticosterone was measured oncebasally in the morning by submandibular venipuncturewithin 1 h of lights-on on day 28 of imipramine treatment(day 58 of defeat) and once two days later, within 4 h oflights-on, by decapitation 30 min after social defeat in allgroups (day 60, after a total of 29 d imipramine treatment).Plasma corticosterone was assayed as previously describedwith a radioimmunoassay kit from MPBiomedical (Solon, OH),

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 119

using all reagents and samples at half-volume (Jacobsonet al., 1997).

4.5. In situ hybridization

In situ hybridization analysis of c-fos mRNA was performedas previously described (Bowens et al., 2012). In brief, hybri-dization was performed at 60 1C in 10 m sections of fresh-frozen brain using a 35S-cRNA probe complementary to the1.5 kb Pst I fragment of the mouse c-fos cDNA (pGEMfos3;Dr. Michael Greenberg, Children's Hospital, Boston, MA (vanBeveren et al., 1983). All other steps were as described byBowens et al. (2012).

4.6. Data analysis

4.6.1. In situ hybridizationDensitometric readings were collected by exposing slides tophosphorimager screens that were scanned at a 50 mresolution (Typhoon 9210, GE Health Care, Niskayuna, NY)and analyzed with ImageQuant 5.0 software (GE Health Care,Niskayuna, NY). Each screen was exposed to slides from micein every treatment group and to a set of identical 14Cstandards (146A and 146B, American Radiolabeled Chemicals,St. Louis, MO) that was used to normalize readings amongscreens. Densitometric readings from hand-drawn outlines ofeach brain region were corrected for background from acorresponding non-expressing area in the same section andscaled to the size of the largest outline for a given region. Atleast two readings were averaged for each region and mouse.

4.6.2. StatisticsData were analyzed by 2- and 3-way ANOVA for Experiments1 and 2, respectively. Post-hoc comparisons where ANOVAmain effects or interactions were significant were made by t-test with Bonferroni correction. Data are presented asmean7SEM; significance was defined as Po0.05.

Acknowledgments

This work was supported by RO1 MH080394 to the author.The author has no conflicts of interest to report.

r e f e r e n c e s

Barik, J., Marti, F., Morel, C., Fernandez, S.P., Lanteri, C., Godeheu,G., Tassin, J.P., Mombereau, C., Faure, P., Tronche, F., 2013.Chronic stress triggers social aversion via glucocorticoidreceptor in dopaminoceptive neurons. Science 339, 332335.

Berton, O., McClung, C.A., DiLeone, R.J., Krishnan, V., Renthal, W.,Russo, S.J., Graham, D., Tsankova, N.M., Bolanos, C.A., Rios,M., Monteggia, L.M., Self, D.W., Nestler, E.J., 2006. Essential roleof BDNF in the mesolimbic dopamine pathway in social defeatstress. Science 311, 864868.

van Beveren, C., van Straaten, F., Curran, T., Muller, R., Verma, I.M., 1983. Analysis of FBJ-MuSV provirus and c-fos (mouse)gene reveals that viral and cellular fos gene products havedifferent carboxy termini. Cell 32, 12411255.

Bhatnagar, S., Huber, R., Nowak, N., Trotter, P., 2002. Lesions of theposterior paraventricular thalamus block habituation of

hypothalamicpituitaryadrenal responses to repeatedrestraint. J. Neuroendocrinol. 14, 403410.

Bhatnagar, S., Huber, R., Lazar, E., Pych, L., Vining, C., 2003.Chronic stress alters behavior in the conditioned defensiveburying test: role of the posterior paraventricular thalamus.Pharmacol. Biochem. Behav. 76, 343349.

Blugeot, A., Rivat, C., Bouvier, E., Molet, J., Mouchard, A., Zeau, B.,Bernard, C., Benoliel, J.J., Becker, C., 2011. Vulnerability todepression: from brain neuroplasticity to identification ofbiomarkers. J. Neurosci. 31, 1288912899.

Bowens, N., Heydendael, W., Bhatnagar, S., Jacobson, L., 2012.Lack of elevations in glucocorticoids correlates withdysphoria-like behavior after repeated social defeat. Physiol.Behav. 105, 958965.

Boyle, M.P., Brewer, J.A., Funatsu, M., Wozniak, D.F., Tsien, J.Z.,Izumi, Y., Muglia, L.J., 2005. Acquired deficit of forebrainglucocorticoid receptor produces depression-like changes inadrenal axis regulation and behavior. Proc. Natl. Acad. Sci.USA 102, 473478.

Boyle, M.P., Kolber, B.J., Vogt, S.K., Wozniak, D.F., Muglia, L.J., 2006.Forebrain glucocorticoid receptors modulate anxiety-associated locomotor activation and adrenal responsiveness.J. Neurosci. 26, 19711978.

Brewer, J.A., Khor, B., Vogt, S.K., Muglia, L.M., Fujiwara, H.,Haegele, K.E., Sleckman, B.P., Muglia, L.J., 2003. T-cellglucocorticoid receptor is required to suppress COX-2-mediated lethal immune activation. Nat. Med. 9, 13181322.

Brown, E.S., 2009. Effects of glucocorticoids on mood, memory,and the hippocampus. Treatment and preventive therapy.Ann. N. Y. Acad. Sci. 1179, 4155.

Calfa, G., Bussolino, D., Molina, V.A., 2007. Involvement of thelateral septum and the ventral hippocampus in the emotionalsequelae induced by social defeat: role of glucocorticoidreceptors. Behav. Brain Res. 181, 2334.

Caspi, A., Sugden, K., Moffitt, T.E., Taylor, A., Craig, I.W.,Harrington, H., McClay, J., Mill, J., Martin, J., Braithwaite, A.,Poulton, R., 2003. Influence of life stress on depression:moderation by a polymorphism in the 5-HTT gene. Science301, 386389.

Cryan, J.F., Markou, A., Lucki, I., 2002. Assessing antidepressantactivity in rodents: recent developments and future needs.Trends Pharmacol. Sci. 23, 238245.

Der-Avakian, A., Mazei-Robison, M.S., Kesby, J.P., Nestler, E.J.,Markou, A., 2014. Enduring deficits in brain reward functionafter chronic social defeat in rats: susceptibility, resilience,and antidepressant response. Biol. Psychiatry http://dxdoi.org/10.1016/j.biopsych.2014.01.013 (in press).

Van Dijken, H.H., Mos, J., van der Heyden, A.M., Tilders, F.J.H.,1992. Characterization of stress-induced long-term behavioralchanges in rats: evidence in favor of anxiety. Physiol. Behav.52, 945951.

Dubreucq, S., Matias, I., Cardinal, P., Haring, M., Lutz, B.,Marsicano, G., Chaouloff, F., 2012. Genetic dissection of therole of cannabinoid type-1 receptors in the emotionalconsequences of repeated social stress in mice.Neuropsychopharmacology 37, 18851900.

Fanselow, M.S., Dong, H.W., 2010. Are the dorsal and ventralhippocampus functionally distinct structures?. Neuron65, 719.

Felix-Ortiz, A.C., Beyeler, A., Seo, C., Leppla, C.A., Wildes, C.P., Tye,K.M., 2013. BLA to vHPC inputs modulate anxiety-relatedbehaviors. Neuron 79, 658664.

File, S.E., 1980. The use of social interaction as a method fordetecting anxiolytic activity of chlordiazepoxide-like drugs.J. Neurosci. Methods 2, 219238.

Furay, A.R., Bruestle, A.E., Herman, J.P., 2008. The role of theforebrain glucocorticoid receptor in acute and chronic stress.Endocrinology 149, 54825490.

http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref1http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref1http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref1http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref1http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref2http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref2http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref2http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref2http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref2http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref3http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref3http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref3http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref3http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref4http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref4http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref4http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref4http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref5http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref5http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref5http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref5http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref6http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref6http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref6http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref6http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref7http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref7http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref7http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref7http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref8http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref8http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref8http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref8http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref8http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref9http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref9http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref9http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref9http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref10http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref10http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref10http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref10http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref11http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref11http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref11http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref12http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref12http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref12http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref12http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref13http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref13http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref13http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref13http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref13http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref14http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref14http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref14http://dx.doi.org/10.1016/j.biopsych.2014.01.013http://dx.doi.org/10.1016/j.biopsych.2014.01.013http://dx.doi.org/10.1016/j.biopsych.2014.01.013http://dx.doi.org/10.1016/j.biopsych.2014.01.013http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref16http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref16http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref16http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref16http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref17http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref17http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref17http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref17http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref17http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref18http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref18http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref18http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref19http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref19http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref19http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref20http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref20http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref20http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref22http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref22http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref22

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1120

Ginsberg, A.B., Campeau, S., Day, H.E., Spencer, R.L., 2003. Acuteglucocorticoid pretreatment suppresses stress-inducedhypothalamicpituitaryadrenal axis hormone secretion andexpression of corticotropin-releasing hormone hnRNA butdoes not affect c-fos mRNA or Fos protein expression in theparaventricular nucleus of the hypothalamus. J.Neuroendocrinol. 15, 10751083.

Institute of Laboratory Animal Resources, Commission on LifeSciences, National Research Council, 2010. Guide for the careand use of laboratory animals. National Academies Press,Washington, DC.

Ising, M., Horstmann, S., Kloiber, S., Lucae, S., Binder, E.B.,Kern, N., Kunze, H.E., Pfennig, A., Uhr, M., Holsboer, F., 2007.Combined dexamethasone/corticotropin releasing hormonetest predicts treatment response in major depression apotential biomarker?. Biol. Psychiatry 62, 4754.

Jacobs, S., Hansen, F., Kasl, S., Ostfeld, A., Berkman, L., Kim, K.,1990. Anxiety disorders during acute bereavement: risk andrisk factors. J. Clin. Psychiatry 51, 269274.

Jacobson, L., 2005. Hypothalamicpituitaryadrenocortical axisregulation. Endocrinol. Metab. Clin. N. Am. 34, 271292.

Jacobson, L., Zurakowski, D., Majzoub, J.A., 1997. Proteinmalnutrition increases plasma ACTH and anterior pituitarypro-opiomelanocortin messenger ribonucleic acid in the rat.Endocrinology 138, 10481057.

Johansen, J.P., Cain, C.K., Ostroff, L.E., LeDoux, J.E., 2011. Molecularmechanisms of fear learning and memory. Cell 147, 509524.

Kenna, H.A., Poon, A.W., de los Angeles, C.P., Koran, L.M., 2011.Psychiatric complications of treatment with corticosteroids:review with case report. Psychiatry Clin. Neurosci. 65, 549560.

Kolber, B.J., Roberts, M.S., Howell, M.P., Wozniak, D.F., Sands, M.S.,Muglia, L.J., 2008. Central amygdala glucocorticoid receptoraction promotes fear-associated CRH activation andconditioning. Proc. Natl. Acad. Sci. USA 105, 1200412009.

Kollack-Walker, S., Don, C., Watson, S.J., Akil, H., 1999.Differential expression of c-fos mRNA within neurocircuits ofmale hamsters exposed to acute or chronic defeat.J. Neuroendocrinol. 11, 547559.

Korte, S.M., De Kloet, E.R., Buwalda, B., Bouman, S.D., Bohus, B.,1996. Antisense to the glucocorticoid receptor in hippocampaldentate gyrus reduces immobility in the forced swim test. Eur.J. Pharmacol. 301, 1925.

Krishnan, V., Han, M.H., Graham, D.L., Berton, O., Renthal, W.,Russo, S.J., Laplant, Q., Graham, A., Lutter, M., Lagace, D.C.,Ghose, S., Reister, R., Tannous, P., Green, T.A., Neve, R.L.,Chakravarty, S., Kumar, A., Eisch, A.J., Self, D.W., Lee, F.S.,Tamminga, C.A., Cooper, D.C., Gershenfeld, H.K., Nestler, E.J.,2007. Molecular adaptations underlying susceptibility andresistance to social defeat in brain reward regions. Cell 131,391404.

Lai, M., Horsburgh, K., Bae, S.E., Carter, R.N., Stenvers, D.J., Fowler,J.H., Yau, J.L., Gomez-Sanchez, C.E., Holmes, M.C., Kenyon, C.J.,Seckl, J.R., Macleod, M.R., 2007. Forebrain mineralocorticoidreceptor overexpression enhances memory, reduces anxietyand attenuates neuronal loss in cerebral ischaemia. Eur.J. Neurosci. 25, 18321842.

Maes, M., Mylle, J., Delmeire, L., Altamura, C., 2000. Psychiatricmorbidity and comorbidity following accidental man-madetraumatic events: incidence and risk factors. Eur. Arch.Psychiatry Clin. Neurosci. 250, 156162.

Markham, C.M., Taylor, S.L., Huhman, K.L., 2010. Role ofamygdala and hippocampus in the neural circuit subservingconditioned defeat in Syrian hamsters. Learn. Mem. 17,109116.

McKlveen, J.M., Myers, B., Flak, J.N., Bundzikova, J., Solomon, M.B.,Seroogy, K.B., Herman, J.P., 2013. Role of prefrontal cortexglucocorticoid receptors in stress and emotion. Biol.Psychiatry 74, 672679.

Mitchell, J.B., Meaney, M.J., 1991. Effects of corticosterone onresponse consolidation and retrieval in the forced swim test.Behav. Neurosci. 105, 798803.

Montkowski, A., Barden, N., Wotjak, C., Stec, I., Ganster, J.,Meaney, M., Engelman, M., J.M.H.M., Reul, Landgraf, R.,Holsboer, F., 1995. Long-term antidepressant treatmentreduces behavioral deficits in transgenic mice with impairedglucocorticoid receptor function. J. Neuroendocrinol. 7,841845.

Mostalac-Preciado, C.R., deGortari, P., Lopez-Rubalcava, C., 2011.Antidepressant-like effects of mineralocorticoid but notglucocorticoid antagonists in the lateral septum: interactionswith the serotonergic system. Behav. Brain Res. 223, 8898.

Nestler, E.J., Carlezon Jr., W.A., 2006. The mesolimbic dopaminereward circuit in depression. Biol. Psychiatry 59, 11511159.

Nikulina, E.M., Covington 3rd, H.E., Ganschow, L., Hammer Jr, R.P.,Miczek, K.A., 2004. Long-term behavioral and neuronal cross-sensitization to amphetamine induced by repeated briefsocial defeat stress: Fos in the ventral tegmental area andamygdala. Neuroscience 123, 857865.

Platt, J.E., Stone, E.A., 1982. Chronic restraint stress elicits apositive antidepressant response on the forced swim test. Eur.J. Pharmacol. 82, 179181.

Porsolt, R.D., Bertin, A., Jalfre, M., 1977a. Behavioral despair inmice: a primary screening test for antidepressants. Arch. Int.Pharmacodyn. 229, 327336.

Porsolt, R.D., LePichon, M., Jalfre, M., 1977b. Depression: a newanimal model sensitive to antidepressant treatments. Nature266, 730732.

Rakofsky, J.J., Holtzheimer, P.E., Nemeroff, C.B., 2009. Emergingtargets for antidepressant therapies. Curr. Opin. Chem. Biol.13, 291302.

Razzoli, M., Carboni, L., Andreoli, M., Michielin, F., Ballottari, A.,Arban, R., 2011. Strain-specific outcomes of repeated socialdefeat and chronic fluoxetine treatment in the mouse.Pharmacol. Biochem. Behav. 97, 566576.

Rozeboom, A.M., Akil, H., Seasholtz, A.F., 2007. Mineralocorticoidreceptor overexpression in forebrain decreases anxiety-likebehavior and alters the stress response in mice. Proc. Natl.Acad. Sci. 104, 46884693.

Schelling, G., Roozendaal, B., Krauseneck, T., Schmoelz, M.,de Quervain, D., Briegel, J., 2006. Efficacy of hydrocortisone inpreventing posttraumatic stress disorder following criticalillness and major surgery. Ann. N. Y. Acad. Sci. 1071, 4653.

Smythe, J.W., Murphy, D., Timothy, C., Costall, B., 1997.Hippocampal mineralocorticoid, but not glucocorticoid,receptors modulate anxiety-like behavior in rats. Pharmacol.Biochem. Behav. 56, 507513.

Sotres-Bayon, F., Sierra-Mercado, D., Pardilla-Delgado, E.,Quirk, G.J., 2012. Gating of fear in prelimbic cortex byhippocampal and amygdala inputs. Neuron 76, 804812.

Sterner, E.Y., Kalynchuk, L.E., 2010. Behavioral andneurobiological consequences of prolonged glucocorticoidexposure in rats: relevance to depression. Prog.Neuropsychopharmacol. Biol. Psychiatry 34, 777790.

Sudheimer, K.D., Abelson, J.L., Taylor, S.F., Martis, B., Welsh, R.C.,Warner, C., Samet, M., Manduzzi, A., Liberzon, I., 2013.Exogenous glucocorticoids decrease subgenual cingulateactivity evoked by sadness. Neuropsychopharmacology 38,826845.

Swiergiel, A.H., Zhou, Y., Dunn, A.J., 2007. Effects of chronicfootshock, restraint and corticotropin-releasing factor onfreezing, ultrasonic vocalization and forced swim behavior inrats. Behav. Brain Res. 183, 178187.

Tronche, F., Kellendonk, C., Kretz, O., Gass, P., Anlag, K.,Orban, P.C., Bock, R., Klein, R., Schutz, G., 1999. Disruption ofthe glucocorticoid receptor gene in the nervous sytem resultsin reduced anxiety. Nat. Genet. 23, 99103.

http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref23http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref21http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref21http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref21http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref21http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref24http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref24http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref24http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref24http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref24http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref25http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref25http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref25http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref26http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref26http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref27http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref27http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref27http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref27http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref28http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref28http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref29http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref29http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref29http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref30http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref30http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref30http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref30http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref31http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref31http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref31http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref31http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref32http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref32http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref32http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref32http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref33http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref34http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref35http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref35http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref35http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref35http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref36http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref36http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref36http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref36http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref37http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref37http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref37http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref37http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref38http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref38http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref38http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref39http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref40http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref40http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref40http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref40http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref41http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref41http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref42http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref42http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref42http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref42http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref42http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref43http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref43http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref43http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref44http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref44http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref44http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref45http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref45http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref45http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref46http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref46http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref46http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref47http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref47http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref47http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref47http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref48http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref48http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref48http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref48http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref49http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref49http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref49http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref49http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref50http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref50http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref50http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref50http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref51http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref51http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref51http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref52http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref52http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref52http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref52http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref53http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref53http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref53http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref53http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref53http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref54http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref54http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref54http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref54http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref55http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref55http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref55http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref55

b r a i n r e s e a r c h 1 5 8 3 ( 2 0 1 4 ) 1 0 9 1 2 1 121

Tsien, J.Z., Chen, D.F., Anderson, D.J., Mayford, M., Kandel, E.R.,Tonegawa, S., 1996. Subregion- and cell type-restricted geneknockout in mouse brain. Cell 87, 13171326.

Veldhuis, H.D., C.C.M.M., De Korte, De Kloet, E.R., 1985.Glucocorticoids facilitate the retention of acquired immobilityduring forced swimming. Eur. J. Pharmacol. 115, 211217.

Ver Hoeve, E.S., Kelly, G., Luz, S., Ghanshani, S., Bhatnagar, S.,2013. Short-term and long-term effects of repeated socialdefeat during adolescence or adulthood in female rats.Neuroscience 249, 6373.

Vincent, M., Hussain, R., Zampi, M., Sheeran, K., Solomon, M.B.,Herman, J.P., Jacobson, L., 2013. Sensitivity of depression-likebehavior to glucocorticoids and antidepressants isindependent of forebrain glucocorticoid receptors. Brain Res.1525, 115.

Wei, Q., Lu, X.-Y., Liu, L., Schafer, G., Shieh, K.-R., Burke, S.,Robinson, T.E., Watson, S.J., Seasholtz, A.F., Akil, H., 2004.Glucocorticoid receptor overexpression in forebrain: a mousemodel of increased emotional lability. Proc. Natl. Acad. Sci.USA 101, 1185111856.

Weinberg, M.S., Johnson, D.C., Bhatt, A.P., Spencer, R.L., 2010.

Medial prefrontal cortex activity can disrupt the expression of

stress response habituation. Neuroscience 168, 744756.Willner, P., 2005. Chronic mild stress (CMS) revisited: consistency

and behavioural-neurobiological concordance in the effects of

CMS. Neuropsychobiology 52, 90110.Wolkowitz, O.M., Burke, H., Epel, E.S., Reus, V.I., 2009.

Glucocorticoids. Mood, memory, and mechanisms. Ann. N. Y.

Acad. Sci. 1179, 1940.Wu, T.C., Chen, H.T., Chang, H.Y., Yang, C.Y., Hsiao, M.C., Cheng,

M.L., Chen, J.C., 2013. Mineralocorticoid receptor antagonist

spironolactone prevents chronic corticosterone induced

depression-like behavior. Psychoneuroendocrinology 38,

871883.Yang, Y.L., Chao, P.K., Lu, K.T., 2006. Systemic and intra-amygdala

administration of glucocorticoid agonist and antagonist

modulate extinction of conditioned fear.

Neuropsychopharmacology 31, 912924.

http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref56http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref56http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref56http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref57http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref57http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref57http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref58http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref58http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref58http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref58http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref59http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref59http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref59http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref59http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref59http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref60http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref60http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref60http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref60http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref60http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref61http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref61http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref61http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref62http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref62http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref62http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref63http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref63http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref63http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref64http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref64http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref64http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref64http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref64http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref65http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref65http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref65http://refhub.elsevier.com/S0006-8993(14)01130-5/sbref65Forebrain glucocorticoid receptor gene deletion attenuates behavioral changes and antidepressant responsiveness during...IntroductionResultsExperiment 1: effects of forebrain GR deletion on behavioral and HPA responses to repeated social defeatExperiment 2: effects of forebrain GR deletion on behavioral and HPA responses to antidepressant treatment during...DiscussionExperimental proceduresAnimalsExperimentsBehavioral testsSocial interaction testSucrose preferenceForced swim testTail suspension testPlasma corticosterone assayIn situ hybridizationData analysisIn situ hybridizationStatisticsAcknowledgmentsReferences