Adaptive Control of 5-HT Dorsal Raphe by 5-HT4 in the Prefrontal Cortex Prevents Persistent Hypophagia following Stress

HAL Id: hal-01825927

https://hal.archives-ouvertes.fr/hal-01825927

Submitted on 28 Jun 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Adaptive Control of 5-HT Dorsal Raphe by 5-HT4 in the Prefrontal Cortex Prevents Persistent Hypophagia

following Stress

Alexandra Jean, Laetitia Laurent, Sabira Delaunay, Stéphane Doly, Nicole Dusticier, David Linden, Rachael Neve, Luc Maroteaux, André Nieoullon,

Valerie Compan

To cite this version:

Alexandra Jean, Laetitia Laurent, Sabira Delaunay, Stéphane Doly, Nicole Dusticier, et al.. Adaptive Control of 5-HT Dorsal Raphe by 5-HT4 in the Prefrontal Cortex Prevents Persistent Hypophagia following Stress. Cell Reports, Elsevier Inc, 2017, 21 (4), pp.901 - 909. �10.1016/j.celrep.2017.10.003�.

�hal-01825927�

Adaptive Control of 5-HT Dorsal Raphe by 5-HT

in the Prefrontal Cortex Prevents Persistent Hypophagia Following Stress

Alexandra Jean

, Laetitia Laurent

, Sabira Delaunay

, Stéphane Doly

, Nicole Dusticier

, David R. Linden

, Rachael Neve

, Luc Maroteaux

, André Nieoullon

and Valérie Compan

1

Department of Sciences, Brain & Anorexia, Nîmes University Nîmes 30000 France

2

INSERM UMR S839 Paris 75000, France

3

CNRS UMR 7288 Aix-Marseille University Marseille 13 288 France

4

Department of Physiology & Biomedical Engineering, Mayo Clinic Rochester MN 55905 USA

5

Department of Brain & Cognitive Sciences, Massachusetts Institute of Technology, Cambridge MA 02139-4307 USA

*Lead contact Pr. Valérie Compan

Correspondence valerie.compan@unimes.fr

SUMMARY

Transient reduced food intake (hypophagia) following high stress could have beneficial effects on longevity, but paradoxically, hypophagia can persist and become anorexia-like behavior. The neural underpinnings of stress-induced-hypophagia and how the brain can prevent the transition from transient to persistent hypophagia remain undetermined. Here, we propose the involvement of a network governing goal- directed behavior (decision). This network consists of the ascending serotonergic inputs from the dorsal raphe nucleus (DR) to the medial prefrontal cortex (mPFC). The serotonin 4 receptors (5-HT

Rs) in the mPFC are now reported to be necessary and sufficient for stress-induced hypophagia because their restitution in 5-HT

R knockout adult mice restores this phenotype. The restored brains also regain the ability to control signals of depression resistance in the DR (reduced 5-HT uptake and 5-HT

R levels, 5-HT accumulation).

The DR-5-HT

R counterbalances the mPFC-5-HT

Rs to prevent persistent hypophagia. An early anorexia could then circumvent depression through adaptive responses to stress.

IN BRIEF

Jean et al. report causal relationships between the serotonin 4 receptors and stress-induced hypophagia, attributable

to specific neural signals of depression resistance in the dorsal raphe nucleus, which protect from an early anorexia.

HIGHLIGHTS

=

mPFC-5-HT

Rs are causally linked to hypophagia following stress

=

mPFC-5-HT

Rs mediate stress-induced changes in DR-5-HT parameters

=

mPFC-5-HT

Rs are counterbalanced by DR-5-HT

to prevent an early anorexia

=

Hypophagia due to stress is not implemented during early stage of development

INTRODUCTION

In the face of environmental changes, behavioral disturbances often correlate with deregulations of neural circuits.

Exploring these correlations in simpler model transgenic animals makes possible the study of phenotypes in isolation from molecular to behaviors and has revealed the conservation of specific molecular mechanisms in humans [(Bevilacqua et al., 2010; Kirov et al., 2012), reviewed in (Donaldson and Hen, 2015)].

Food intake is an evolutionarily conserved behavior across all species and involves numerous biological systems including the phylogenetically old serotonergic system. In mammals, the serotonergic neuronal cell bodies assemble in the raphe nuclei [reviewed in (Azmitia, 1999)]. Among nine nuclei, the dorsal and median raphe nuclei (DR, MR) send axons to the whole forebrain [reviewed in (Azmitia, 1999)]. In particular, the serotonergic axons in the cerebral cortex mainly arise from the DR (Muzerelle et al., 2016). Serotonin (5-hydroxytryptamine: 5-HT) binds 18 G-protein coupled receptors (5-HTRs), more often located at 100 µm [volume transmission (Descarries et al., 1975)] than at 20 nm (synaptic transmission) from the site of 5-HT release. The preponderant 5-HT volume transmission extends the ubiquitous distribution of the 5-HT system, supporting their multiple functions.

The 5-HT system commonly mediates reduction in food intake, i.e. hypophagia [reviewed (Compan et al., 2015)].

Stimulating the Gi-coupled 5-HT

receptor in the DR (DR-5-HT

R) reduces the firing activity of DR 5-HT neurons (Figure S1), mediating hyperphagia [reviewed in (Compan, 2013)]. Most studies describe hypophagia following stimulation of the 5-HT

and 5-HT

receptors in the hypothalamus [5-HT

R, 5-HT

R, reviewed (Compan et al., 2015)] whereas the 5-HT

R and 5-HT

R can exceptionally serve to enhance feeding (Yadav et al., 2009). The serotonergic system can also mediate motivation for food in food-deprived mice, mediating anorexia-like behavior through the activation of addictive signaling (cAMP/PKA/CART: cocaine- and amphetamine-regulated transcript) under the control of the 5-HT

Rs in the nucleus accumbens (NAc), a critical structure in the brain’s reward system (Jean et al., 2007; Jean et al., 2012)].

Humans who had recovered from one of the symptoms of anorexia nervosa, i.e. from persistent food restriction (called anorexia), show an elevated activity of the DR-5-HT

R (Bailer et al., 2007). In contrast, 5-HT depletion and compensatory high levels of NAc-5-HT

Rs in both rats and humans are seen in obesity (Compan et al., 1996b;

Haahr et al., 2012; Haahr et al., 2014; Ratner et al., 2012)

Food intake then depends on the activity of the 5-HT system and, both are influenced by external stressors [reviewed in (Compan, 2013; Hardaway et al., 2015)], but whether changes in food intake and the activity of the 5- HT system in the face to external stress are causally related or correlated, remain undetermined. One of the most employed animal models to identify the neural basis of hypophagia due to stress is forced immobilization, called the restraint stress [reviewed in (Laurent et al., 2012)]. Mice can neither escape nor learn how to escape from the stressor. This stressor is out of control [uncontrollable stress (Amat et al., 2005)] and enhances 5-HT transmission [reviewed in (Laurent et al., 2012)]. Restrained mice treated with 8-OH-DPAT (Dourish et al., 1987), a 5-HT

R/5- HT

R agonist, or lacking the 5-HT

Rs exhibit attenuated hypophagia (Compan et al., 2004).

The cerebral location of the mediation of hypophagia due to stress by the 5-HT

Rs is unknown. Their cerebral distribution is conserved in humans, with the highest levels in the NAc and the lowest, in the cerebral cortex (Bonaventure et al., 2000; Compan et al., 1996a). The 5-HT

Rs mediate a positive feedback on the DR-5-HT cells, not from the DR (they are absent) but from the ventral part of the medial prefrontal cortex (mPFC, Figure S1).

Here, we tested whether the 5-HT

Rs in the mPFC (mPFC-5-HT

Rs) are necessary and sufficient to mediate

hypophagia due to stress. The experimental procedures employed here, have been utilized in our earlier studies

(Compan et al., 2004; Jean et al., 2007; Jean et al., 2012). In particular, the expression of the 5-HT

Rs was rescued

in the 5-HT

R knockout (KO) mice by transferring, in the mPFC, the Htr4 gene (Herpès simplex virus: HSV5-

HT

R transduced mice). In parallel, knockdown of the mPFC-5-HT

Rs was induced by small interference RNA

(si5-HT

R). These latter experiments require additional WT mice only and were conducted in WT mice from a

129SvPas similar genetic background (Supplemental Experimental Procedures).

RESULTS

No Differences in Food Intake Per Active Tissue Between 129SvTer and 129SvPas WT Mice

Food intake and body weight were lower in 129SvPas than in 129SvTer WT mice (Figures 1A and 1B), but both strains ate a similar amount of food per active tissue [body weight

(West et al., 1997)] when untreated (Figure 1C) or infused, in the mPFC, with control treatments (siCt, HSVLacZ) in the basal condition (Figure 1D).

Knockdown of the 5-HT

Rs in the mPFC Induces Overeating in the Basal Condition

Injecting si5-HT

R in the mPFC induced overeating at 3 h post-infusion compared with control (siCt, Figure 1D). It reduced the levels of both mRNA (38%, Figure 1E) and binding site (49%, Figures 1G and 1H) of the mPFC-5- HT

Rs at 5 h compared with controls. No modifications in food intake or in the mRNA levels of mPFC-5-HT

Rs were detected at 24 h post-treatments in the mPFC compared with controls (means ± SEM, food intake: siCt 4.19

± 0.16 g, si5-HT

R 4.01 ± 0.19 g; 5-HT

R mRNA/18S x 10

: siCt 2.28 ± 0.31, si5-HT

R 2.10 ± 0.15).

Overexpression of the 5-HT

Rs in the mPFC Triggers Hypophagia in the Basal Condition

HSV5-HT

R-transduced WT mice ate less at 3 h (Figures 1D) until 48 h (Figure 2A: days 3-4) and consumed a normal amount of food at 72 h (Figures 2B: day 5) post-injection compared with control (HSVLacZ) in the basal condition, as seen in HSV5-HT

R-transduced 5-HT

R KO mice at 48 and 72 h (Figures 2A and 2B).

HSV5-HT

R-transduced mice of both genotypes displayed increases in the mRNA and binding site levels of mPFC-5-HT

Rs at 77 h post-injection compared with controls (Figures 1F and, 1G: WT 23%, 5-HT

R KO 783%).

The levels of mPFC-5-HT

Rs were lower in HSV5-HT

R-transduced mutant than in control siCt/HSVLacZ-treated WT mice because the examined tissue in mutant mice is deprived of 5-HT

Rs (Figures 1G: -29% and 1H) and, were higher in HSV5-HT

R-transduced mutant than in si5-HT

R-treated WT mice (39%, Figure 1G).

The mPFC-5-HT

Rs Mediate Hypophagia Due to Stress

We next tested whether there is a causal relationship between the mPFC-5-HT

Rs and hypophagia following stress.

Unstressed 129SvTer WT mice still ate more than unstressed 129SvPas WT mice (Figures 2B and 2D). Stress induced a transient hypophagia in WT mice of both strains but not in HSVLacZ-transduced 5-HT

R KO mice compared with controls (Figure 2B-D). The elective rescue of the mPFC-5-HT

Rs in 5-HT

R KO mice restored stress-induced hypophagia (Figure 2C). Hypophagia was not enhanced in stressed WT mice with mPFC-5-HT

R overexpression (Figure 2B), suggesting adaptive changes and/or a ceiling effect.

The mRNA levels of the mPFC-5-HT

Rs in HSVLacZ-transduced 129SvTer WT were lower than in unstressed siCt-treated 129SvPas WT mice (Figures 2F and 2H), consistently with their difference in basal food intake (Figures 2B and 2D). Similarly, stress-induced hypophagia also provoked increases in the mRNA levels of mPFC- 5-HT

Rs in WT compared with unstressed WT animals in both strains (Figures 2F and 2H). In contrast, stress failed to provoke hypophagia and to increase the mRNA levels of the mPFC-5-HT

Rs following injection of si5- HT

R in the mPFC in 129SvPas WT mice compared with controls (Figures 2D and 2H). The 5-HT

R knockdown in other sites (cingulate cortex area 2, NAc) failed to alter stress-induced hypophagia compared with controls (e.g.

NAc: Figures S2A and S2B). Accordingly, the mRNA levels of NAc-5-HT

Rs were reduced in stressed WT mice and, mPFC-5-HT

R-dependent because it was attenuated in mice with mPFC-5-HT

R knockdown (Figure S2C).

Finally, injection, in the mPFC, of an agonist (BIMU8) or antagonist (RS39604) of the 5-HT

Rs mimicked the feeding responses induced by the knockdown and overexpression of the mPFC-5-HT

Rs in the basal and stressful conditions (Figures S2D and S2E). Additionally, stimulation of the mPFC-5-HT

Rs attenuated the weakness of motor reactivity to novelty (Figure S2F), another classic behavioral response to restraint stress (Kennett et al., 1987). When the mPFC-5-HT

Rs were brought to a standstill, the both unstressed and stressed mice were less active (Figure S2F). Finally, “moving less and eating more” involves the blockade of the mPFC-5-HT

Rs, extending and reinforcing earlier studies (Compan et al., 2004; Jean et al., 2007; Jean et al., 2012). Therefore, locomotion and feeding can independently segregate under the influence of the mPFC-5-HT

Rs, regulating specifically feeding and not all responses to stress.

The mPFC-5-HT

Rs Control DR-5-HT Responses to Stress

We next set out to explore the mechanisms whereby the mPFC-5-HT

Rs mediate hypophagia due to stress.

The abnormal resistance to stress-induced hypophagia in the 5-HT

R KO mice was not supported by a general

maladaptive response to stress (Figures S2G and S2H). We therefore focused on the DR-5-HT system. Because,

recording the DR-5-HT neurons firing in freely moving stressed mice was not possible for us, we evaluated the

mRNA levels of Fos, a marker of neural activity, in the raphe nuclei (DR/MR). The mRNA levels of Fos were

increased in stressed WT mice, but less in 5-HT

R KO (Figure 3A) and si5-HT

R-mPFC-treated WT mice (Figure

S3). Rescuing mPFC-5-HT Rs in mutant mice restored stress-induced increases in the mRNA levels of DR/MR-

Fos (Figure 3A). We then conducted analyses in behaving mice and, detected increases in the levels of extracellular DR-5-HT in stressed WT but not in 5-HT

R KO mice (Figure 3B). The use of si5-HT

R was excluded because it has to be injected 3 h before stress, which overlaps the required 2 h to achieve a steady state with the use of the dialysis probe (Supplemental Procedures). However, stressed HSV5-HT

R-transduced 5-HT

R KO mice mimicked the response of stressed WT mice (Figure 3B).

Nonetheless, stressed mice of both genotypes displayed an increase in the 5-HT turnover index at the end of stress and, to a greater extent in stressed mutant than WT mice (Table S1). The levels of extracellular DR-5-HT in 5- HT

R KO mice could potentially be higher but are likely more rapidly diminished than in controls by a higher uptake (Conductier et al., 2006). Accordingly, the mRNA levels of the 5-HT transporter (SERT) in 5-HT

R KO mice remain high in the stressful compared to the basal conditions while, it was reduced after stress in the DR/MR of WT mice compared to unstressed mice (Figure 3D). Rescuing the mPFC-5-HT

R expression in 5-HT

R KO mice partly restored the adequate response (reduced levels) to stress (Figure 3D). Similar changes in SERT binding site concentration have also been observed in these experimental conditions (not illustrated).

Early Anorexia in Stressed Mice with mPFC-5-HT

R Overexpression and DR-5-HT

R Blockade There were no changes in the mRNA levels of DR-5-HT

R in either group of mice, but stress induced decreases in the levels of receptor binding sites (Figures 3E and 3F). In unstressed HSV5-HT

R-transduced mice of both genotypes, the levels of DR-5-HT

R were increased and, more in 5-HT

R KO than in WT mice (Figure 3F).

Rescuing mPFC-5-HT

R in mutant mice partly restored the response (reduced levels) to stress compared with controls (Figure 3F).

A reduced level in the DR-5-HT

R after stress can favor DR-5-HT cell hyperactivity and hypophagia. Following up homeostasis, stimulation of the DR-5-HT

R should next bring to baseline the activity of DR-5-HT cells, cutting the duration of hypophagia. We have blocked the DR-5-HT

R, with the antagonist WAY100635, the day after stress, and observed a persistent hypophagia in stressed mice with mPFC-5-HT

R overexpression while, neither the DR-5-HT

R blockade nor 5-HT

R overexpression alone maintained the duration of hypophagia (Figures 4A and 4B). Stressed mice displayed body weight loss at 24 h while, only mice treated with mPFC-HSV5-HT

R and DR-5- HT

R antagonist exhibited a loss for 96 h compared with controls (Figures 4C and 4D).

DISCUSSION

The present study describes causal relationships between a molecular network and a stress-dependent food intake.

Stress triggers an elevation in the DR-5-HT release along with reduction of DR-SERT and 5-HT

R levels upon the control of the mPFC-5-HT

Rs, causing a transient hypophagia (Supplemental video). The identified molecular network may protect the brain from an early anorexia-like behavior because blocking the DR-5-HT

R with an overexpression of the mPFC-5-HT

Rs mediates a persistent hypophagia due to stress.

The mPFC-5-HT

Rs are shown here, to be necessary and sufficient for hypophagia due to stress because their restitution in 5-HT

R KO adult mice restores this phenotype, excluding their contribution in other brain areas and developmental processes. The restored brains also regain the ability to accumulate 5-HT and reduce the levels of DR-SERT/5-HT

R after stress, needed for hypophagia due to stress. These neural events are not implemented during earlier developmental stages, but adapt in a rapid flexibility in response to uncontrollable stress.

Among these events exist a possible positive mPFC-5-HT

R control of the DR-5-HT release, consistently with the positive mPFC-5-HT

R feedback (Figure S1). The concomitant changes in the DR-5-HT release and SERT/5- HT

R (summary: Figure 3C) fit with those seen in mice lacking either SERT [DR-5-HT

R reduction (Fabre et al., 2000)], the monoamine oxidase-A (main catabolism enzyme of 5-HT) [5-HT elevation, DR-SERT/5-HT

R desensitization (Evrard et al., 2002)] or, central 5-HT [undetectable 5-HT, 5-HT

R elevation (Araragi and Lesch, 2013), also seen in rats (Compan et al., 1998)]. We suggest a respective 5-HT release-dependent negative and positive mPFC-5-HT

R control of DR-SERT and -5-HT

R. Notably, in stressed 5-HT

R KO mice, the levels of DR-5-HT

R remained at control values while it decreased in stressed WT mice. A suggested positive mPFC-5- HT

Rs tonic control of the DR-5-HT

R (limit a reduction) accords earlier deductions (Amigo et al., 2016;

Conductier et al., 2006).

In addition, the mPFC-5-HT

Rs may relay an earlier increase in the DR-5-HT release because it was restored in stressed mPFC-HSV5-HT

R-transduced 5-HT

R KO mice at 100 and not at 80 min during stress. As we discussed elsewhere (Compan et al., 2004), the hyperactivity of hypothalamo-pituitary adrenal axis could first intervene (Kirby et al., 2000), considering the unchanged levels of corticosterone in stressed mutant mice.

An interesting point is that reduced levels of DR-5-HT

R, increased 5-HT release and stimulation of the mPFC-5-

HT

Rs can limit depressive-like states (Lucas et al., 2007; Richardson-Jones et al., 2010). The neural substrates of a

transient hypophagia due to stress is then included in a neural network of self-preservation (e.g. antidepressant-like

behavior), consistently with the abilities of 5-HT

Rs to limit anhedonia (Amigo et al., 2016) and to favor a

rewarding anorexia (Jean et al., 2007; Jean et al., 2012). Accordingly, the reciprocal 5-HT/GLU neural pathways between the mPFC and the DR (Figure S1A) prevent the implementation of neural basis of depressive-like behavior induced by adverse stressors (Amat et al., 2005; Duman et al., 2016). Here, stress failed to provoke hypophagia when the DR-N-methyl-D-aspartate (NMDA) receptor was blocked (Figure S4A), accordingly with the ability of the mPFC-GLU pathway to increase the DR-5-HT release through NMDA receptor (Celada et al., 2001;

de Kock et al., 2006). There are thus commonalities between adaptive neural responses to stress and the effects of particular antidepressants (SERT blockade, 5-HT accumulation, 5-HT

R-desensitization). In humans, chronic antidepressant treatment induces a 5-HT

R desensitization, hypophagia in rats and reduces bulimia in humans (Jackson et al., 2010; McGuirk et al., 1992).

How the brain supports adapted behavior for escaping adverse effects of stressors (controllable stress) is worth mentioning here because these behavioral strategies, i.e. learning response indicating the end or a means to escape from a stressor, are associated with attenuated DR 5-HT cells activity and the subsequent depression-like behavior (i.e. learned helplessness) provoked by uncontrollable stress (Amat et al., 2005; Grahn et al., 1999; Maswood et al., 1998). Learned helplessness following uncontrollable (and not controllable) stress is absent in animals when the activity of the DR 5-HT cells has been reduced (Maier et al., 1995). When mice escape from stress, the mPFC inhibits stress-induced activation of DR 5-HT cells (Amat et al., 2005). The present conclusion integrates well this scientific context because here, limitation of DR 5-HT cells hyperactivity by the DR-5-HT

R prevents subsequent behavioral pathology, i.e. the transition from a transient to persistent hypophagia owing an uncontrollable stressor.

In line with these results, enhancing the negative 5-HT

R feedback prevents learned helplessness when animals are physically more active (Greenwood et al., 2003), highlighting potentiation of the physical activity on the stress resistance and antidepressant efficiency (Babyak et al., 2000; Salmon, 2001). Although the current study focuses on food intake in an uncontrollable stress condition, analyzing the physical activity of stressed mice treated with 5- HT

R pharmaceuticals was tempting. Mice lacking the 5-HT

Rs in the whole brain, or only in the NAc, are less physically active in the open field (Compan et al., 2004; Jean et al., 2012). Accordingly, mice show also less motor activity when the mPFC-5-HT

Rs are blocked in the basal condition (Figure S2F). Mice are more active in the open field when the NAc-5-HT

Rs are stimulated [or overexpressed, (Jean et al., 2012)] but not when the mPFC-5- HT

Rs are stimulated in unstressed mice. The mPFC-5-HT

Rs could then serve to positively maintain motor reactivity to novelty without enhancing it. Accordingly, reduced motor activity in stressed mice is related to a combined (and physiological) over- and down-expression of the mPFC- and NAc-5-HT

Rs, respectively (Figure S2C). Stimulating the mPFC-5-HT

Rs in stressed mice had only attenuated the weakness of motor reactivity in stressed mice compared with controls (Figure S2F). In sum, when the mPFC-5-HT

Rs are blocked, stressed mice eat more and still move less, thus show all behavioral ingredients to install depressive-like behavior. In contrast, activating the mPFC-5-HT

Rs in stressed mice reduces food intake, favors physical activity following an increase activity of the DR 5-HT neurons accompanied by low levels of DR-SERT/5-HT

R, thereby showing better resistance to stress. Accordingly, SERT or 5-HT

KO mice are respectively less active in home cages and open field (Holmes et al., 2002; Ramboz et al., 1998).

Finally, even though this study demonstrates a causal relationship, how the mPFC-5-HT

Rs interact with GLU and GABA transmission in the mPFC and DR to adapt feeding and energy balance following controllable and/or chronic uncontrollable stress remains to be investigated.

In this paper, neural adaptive responses to stress, known to reduce impaired behavior of self-preservation (depression), initiate a persistent hypophagia following stress. An “early anorexia” could then favor self- preservation via neural pathways concerned with dealing with stress, while obesity often exists with depression [reviewed in (Duman et al., 2016)]. Considering the relevance in certain circumstances of modeling behavioral traits of mental disease [reviewed in (Donaldson and Hen, 2015)], even though rather simplistic by obvious necessity, this study introduces a primary mechanism whereby individuals could shift from transient to persistent food restriction as seen in anorexia nervosa (Walsh, 2013), and make conceivable targeting 5-HT

Rs to treat this incurable disease.

EXPERIMENTAL PROCEDURES Animals

Maintenance and experiments were conducted with male 129SvPas WT, 129SvTer 5-HT

R KO and WT mice [4-6 months old, (Compan et al., 2004)] under standard conditions consistent with The Guide for Care and Use of Laboratory Animals ( MENESR agreements n°D3417213 and n°34-408, authorization n°00905.01) described in the Supplemental Experimental Procedures.

Microsurgery

Stainless guide cannula was implanted and fixed in a specific brain area (mPFC, NAc, DR), described in the

Supplemental Experimental Procedures and elsewhere (Jean et al., 2007) 48 h before the onset of the habituation

period of any experiment. A connected stainless cannula to a microsyringe nanopump was inserted in the guide and each compound was infused through a micro-catheter for 1 min at a rate of 1 µl/min in freely moving animals.

Nucleic Acid Treatments

Details about 5-HT

Rs overexpression and knockdown strategies were previously established (Jean et al., 2012) and are described in the Supplemental Experimental Procedures.

Quantitative Real Time PCR

Brains from stressed and unstressed mice of both genotypes were removed 5 and 77 h after respective injection of siRNA and HSV, freeze and dissect at -20°C of (mPFC, 1.2 mm

) to treat total mRNA and cDNA in reactions containing 5-HT

R primers (Jean et al., 2007) and given in the Supplemental Experimental Procedures.

Biochemical Analyses

The Supplemental Experimental Procedures include details respectively reported in (Compan et al., 1998; Compan et al., 2004) and (Doly et al., 2008; Dusticier and Nieoullon, 1987) to (i) label the 5-HT

R and 5-HT

R on frontal brain sections and (ii) evaluate the levels of tissue 5-HT and metabolite in the DR/MR or, extracellular 5-HT in the DR of unstressed and stressed mice of both genotypes.

Feeding Test

As detailed in Supplemental Information, experiments include three periods: the baseline, the day of treatments and restraint stress and the recovery. An initial handling for weighing (t = 0 min) precedes the injection of si5-HT

R or siCt (50 ng/µl), BIMU8 or RS39604 (40 ng/µl) and NaCl (9‰) in the mPFC at t = 10 min and, the onset of stress at t = 3 h for 110 min (total duration: 5 h). The injection of HSV5-HT

R or HSVLacZ (10

infectious units/ml) in the mPFC was performed 3 days and 3 h before stress and, was or not combined with infusion in the DR of either WAY100635 (45 ng/µl, Day 6: 19 h after the end of the stress period) or MK801 (0.5 ng/µl) immediately after stress and, NaCl (9‰).

Statistical Analysis

Data, presented as mean ± SEM, obtained in multiple sessions over time (food intake) were analyzed using two- way repeated measures analysis of variance (ANOVA, STATVIEW 5 software, SAS Institute Inc., San Francisco, CA, USA). When effects of independent variables (treatment, genotype, time, stress), or interactions were significant, one-way ANOVAs (treatment, genotype, time or stress) were performed. For multiple comparisons, the Scheffé F-test was used. Differences with p < 0.05 were considered significant.

AUTHOR CONTRIBUTIONS

A.J., L.L., S.D., S.D., N.D. and R.N. performed experiments. A.J. produced initial version of figures. S.D., D.L., L.M., R.N. and A.N. assisted with interpretation. R.N. and D.L. assisted with writing. V.C. conceived and designed the overall work, interpreted data, wrote the manuscript, produced figures and, the movie with the assistance of N.

Scarpa and L. Janondy (BINOME®, France).

ACKNOWLEDGEMENTS

We are grateful to H. R. Kissileff for helpful discussion and editing, G. Knudsen, C. Ratner and Y. Charnay for binding studies, L. Forichon and F. Arnal for mouse breeding, T. Vallejos for editing and to ANR, ADOR (Anorexia, Dependence, Obesity & Receptors) Foundation and FRM.

REFERENCES

Amat, J., Baratta, M.V., Paul, E., Bland, S.T., Watkins, L.R., and Maier, S.F. (2005). Medial prefrontal cortex determines how stressor controllability affects behavior and dorsal raphe nucleus. Nat Neurosci 8, 365-371.

Amigo, J., Diaz, A., Pilar-Cuellar, F., Vidal, R., Martin, A., Compan, V., Pazos, A., and Castro, E. (2016). The absence of 5- HT4 receptors modulates depression- and anxiety-like responses and influences the response of fluoxetine in olfactory bulbectomised mice: Adaptive changes in hippocampal neuroplasticity markers and 5-HT1A autoreceptor.

Neuropharmacology 111, 47-58.

Araragi, N., and Lesch, K.P. (2013). Serotonin (5-HT) in the regulation of depression-related emotionality: insight from 5-HT transporter and tryptophan hydroxylase-2 knockout mouse models. Curr Drug Targets 14, 549-570.

Azmitia, E.C. (1999). Serotonin neurons, neuroplasticity, and homeostasis of neural tissue. Neuropsychopharmacology 21, 33S-45S.

Babyak, M., Blumenthal, J.A., Herman, S., Khatri, P., Doraiswamy, M., Moore, K., Craighead, W.E., Baldewicz, T.T., and Krishnan, K.R. (2000). Exercise treatment for major depression: maintenance of therapeutic benefit at 10 months. Psychosom Med 62, 633-638.

Bailer, U.F., Frank, G.K., Henry, S.E., Price, J.C., Meltzer, C.C., Mathis, C.A., Wagner, A., Thornton, L., Hoge, J., Ziolko, S.K., et al. (2007). Exaggerated 5-HT1A but normal 5-HT2A receptor activity in individuals ill with anorexia nervosa. Biol Psychiatry 61, 1090-1099.

Bevilacqua, L., Doly, S., Kaprio, J., Yuan, Q., Tikkanen, R., Paunio, T., Zhou, Z., Wedenoja, J., Maroteaux, L., Diaz, S., et al.

(2010). A population-specific HTR2B stop codon predisposes to severe impulsivity. Nature 468, 1061-1066.

Bonaventure, P., Hall, H., Gommeren, W., Cras, P., Langlois, X., Jurzak, M., and Leysen, J.E. (2000). Mapping of serotonin 5- HT(4) receptor mRNA and ligand binding sites in the post-mortem human brain. Synapse 36, 35-46.

Celada, P., Puig, M.V., Casanovas, J.M., Guillazo, G., and Artigas, F. (2001). Control of dorsal raphe serotonergic neurons by the medial prefrontal cortex: Involvement of serotonin-1A, GABA(A), and glutamate receptors. J Neurosci 21, 9917-9929.

Compan, V. (2013). Under- to overeating: what role do serotonin receptors play? Future Neurology Invited focus article.

Compan, V., Daszuta, A., Salin, P., Sebben, M., Bockaert, J., and Dumuis, A. (1996a). Lesion study of the distribution of serotonin 5-HT4 receptors in rat basal ganglia and hippocampus. Eur J Neurosci 8, 2591-2598.

Compan, V., Dusticier, N., Nieoullon, A., and Daszuta, A. (1996b). Opposite changes in striatal neuropeptide Y immunoreactivity after partial and complete serotonergic depletion in the rat. Synapse 24, 87-96.

Compan, V., Segu, L., Buhot, M.C., and Daszuta, A. (1998). Differential effects of serotonin (5-HT) lesions and synthesis blockade on neuropeptide-Y immunoreactivity and 5-HT1A, 5-HT1B/1D and 5-HT2A/2C receptor binding sites in the rat cerebral cortex. Brain Res 795, 264-276.

Compan, V., Walsh, B.T., Kaye, W., and Geliebter, A. (2015). How Does the Brain Implement Adaptive Decision Making to Eat? J Neurosci 35, 13868-13878.

Compan, V., Zhou, M., Grailhe, R., Gazzara, R.A., Martin, R., Gingrich, J., Dumuis, A., Brunner, D., Bockaert, J., and Hen, R.

(2004). Attenuated response to stress and novelty and hypersensitivity to seizures in 5-HT4 receptor knock-out mice. J Neurosci 24, 412-419.

Conductier, G., Dusticier, N., Lucas, G., Cote, F., Debonnel, G., Daszuta, A., Dumuis, A., Nieoullon, A., Hen, R., Bockaert, J.,

de Kock, C.P., Cornelisse, L.N., Burnashev, N., Lodder, J.C., Timmerman, A.J., Couey, J.J., Mansvelder, H.D., and Brussaard, A.B. (2006). NMDA receptors trigger neurosecretion of 5-HT within dorsal raphe nucleus of the rat in the absence of action potential firing. J Physiol 577, 891-905.

Descarries, L., Beaudet, A., and Watkins, K.C. (1975). Serotonin nerve terminals in adult rat neocortex. Brain Res 100, 563- 588.

Doly, S., Valjent, E., Setola, V., Callebert, J., Herve, D., Launay, J.M., and Maroteaux, L. (2008). Serotonin 5-HT2B receptors are required for 3,4-methylenedioxymethamphetamine-induced hyperlocomotion and 5-HT release in vivo and in vitro. J Neurosci 28, 2933-2940.

Donaldson, Z.R., and Hen, R. (2015). From psychiatric disorders to animal models: a bidirectional and dimensional approach.

Biol Psychiatry 77, 15-21.

Dourish, C.T., Kennett, G.A., and Curzon, G. (1987). The 5-HT1A agonists 8-OH-DPAT, buspirone and ipsapirone attenuate

stress-induced anorexia in rats. J Psychopharmacol 1, 23-30.

Duman, R.S., Aghajanian, G.K., Sanacora, G., and Krystal, J.H. (2016). Synaptic plasticity and depression: new insights from stress and rapid-acting antidepressants. Nat Med 22, 238-249.

Dusticier, N., and Nieoullon, A. (1987). Comparative analysis of the effects of in vivo electrical stimulation of the frontal cortex and g-butyrolactone administration on dopamine and dihydroxyphenylacetic acid (DOPAC) striatal contents in the rat.

Neurochem Int 10, 275-280.

Evrard, A., Malagie, I., Laporte, A.M., Boni, C., Hanoun, N., Trillat, A.C., Seif, I., De Maeyer, E., Gardier, A., Hamon, M., et

Fabre, V., Beaufour, C., Evrard, A., Rioux, A., Hanoun, N., Lesch, K.P., Murphy, D.L., Lanfumey, L., Hamon, M., and Martres, M.P. (2000). Altered expression and functions of serotonin 5-HT1A and 5-HT1B receptors in knock-out mice lacking the 5-HT transporter. Eur J Neurosci 12, 2299-2310.

Grahn, R.E., Will, M.J., Hammack, S.E., Maswood, S., McQueen, M.B., Watkins, L.R., and Maier, S.F. (1999). Activation of serotonin-immunoreactive cells in the dorsal raphe nucleus in rats exposed to an uncontrollable stressor. Brain Res 826, 35-43.

Greenwood, B.N., Foley, T.E., Day, H.E., Campisi, J., Hammack, S.H., Campeau, S., Maier, S.F., and Fleshner, M. (2003).

Freewheel running prevents learned helplessness/behavioral depression: role of dorsal raphe serotonergic neurons. J Neurosci

Haahr, M., Rasmussen, P., Madsen, K., Marner, L., Ratner, C., Gillings, N., Baaré, W., and Knudsen, G. (2012). Obesity is associated with high serotonin 4 receptor availability in the brain reward circuitry. Neuroimage 16, 61(4):, 884-888.

Haahr, M.E., Fisher, P.M., Jensen, C.G., Frokjaer, V.G., Mahon, B.M., Madsen, K., Baare, W.F., Lehel, S., Norremolle, A., Rabiner, E.A., et al. (2014). Central 5-HT4 receptor binding as biomarker of serotonergic tonus in humans: a [11C]SB207145 PET study. Mol Psychiatry 19, 427-432.

Hardaway, J.A., Crowley, N.A., Bulik, C.M., and Kash, T.L. (2015). Integrated circuits and molecular components for stress and feeding: implications for eating disorders. Gene, Brain and Behavior 14, 85-97.

Holmes, A., Murphy, D.L., and Crawley, J.N. (2002). Reduced aggression in mice lacking the serotonin transporter.

Psychopharmacology (Berl) 161, 160-167.

Jackson, C.W., Cates, M., and Lorenz, R. (2010). Pharmacotherapy of eating disorders. Nutr Clin Pract 25, 143-159.

Jean, A., Conductier, G., Manrique, C., Bouras, C., Berta, P., Hen, R., Charnay, Y., Bockaert, J., and Compan, V. (2007).

Anorexia induced by activation of serotonin 5-HT4 receptors is mediated by increases in CART in the nucleus accumbens.

Proc Natl Acad Sci U S A 104, 16335-16340.

Jean, A., Laurent, L., Bockaert, J., Charnay, Y., Dusticier, N., Nieoullon, A., Barrot, M., Neve, R., and Compan, V. (2012).

The nucleus accumbens 5-HTR(4)-CART pathway ties anorexia to hyperactivity. Transl Psychiatry 2, e203.

Kennett, G.A., Dourish, C.T., and Curzon, G. (1987). Antidepressant-like action of 5-HT1A agonists and conventional antidepressants in an animal model of depression. Eur J Pharmacol 134, 265-274.

Kirby, L.G., Rice, K.C., and Valentino, R.J. (2000). Effects of corticotropin-releasing factor on neuronal activity in the serotonergic dorsal raphe nucleus. Neuropsychopharmacology 22, 148-162.

Kirov, G., Pocklington, A.J., Holmans, P., Ivanov, D., Ikeda, M., Ruderfer, D., Moran, J., Chambert, K., Toncheva, D., Georgieva, L., et al. (2012). De novo CNV analysis implicates specific abnormalities of postsynaptic signalling complexes in the pathogenesis of schizophrenia. Mol Psychiatry 17, 142-153.

Laurent, L., Jean, A., Manrique, C., Najimi, M., Chigr, F., and Compan, V. (2012). Anorexia and drugs of abuse abnormally suppress appetite, the result of a shared molecular signal foul-up. . In Animal model of eating disorders N. Avena, ed. (New- York (USA): Springer Book, Humana Press).

Lucas, G., Rymar, V.V., Du, J., Mnie-Filali, O., Bisgaard, C., Manta, S., Lambas-Senas, L., Wiborg, O., Haddjeri, N., Pineyro, G., et al. (2007). Serotonin(4) (5-HT(4)) receptor agonists are putative antidepressants with a rapid onset of action. Neuron 55, 712-725.

Maier, S.F., Grahn, R.E., and Watkins, L.R. (1995). 8-OH-DPAT microinjected in the region of the dorsal raphe nucleus

blocks and reverses the enhancement of fear conditioning and interference with escape produced by exposure to inescapable

shock. Behav Neurosci 109, 404-412.

Maswood, S., Barter, J.E., Watkins, L.R., and Maier, S.F. (1998). Exposure to inescapable but not escapable shock increases extracellular levels of 5-HT in the dorsal raphe nucleus of the rat. Brain Res 783, 115-120.

McGuirk, J., Muscat, R., and Willner, P. (1992). Effects of chronically administered fluoxetine and fenfluramine on food intake, body weight and the behavioural satiety sequence. Psychopharmacology (Berl) 106, 401-407.

Muzerelle, A., Scotto-Lomassese, S., Bernard, J.F., Soiza-Reilly, M., and Gaspar, P. (2016). Conditional anterograde tracing reveals distinct targeting of individual serotonin cell groups (B5-B9) to the forebrain and brainstem. Brain Struct Funct 221, 535-561.

Ramboz, S., Oosting, R., Amara, D.A., Kung, H.F., Blier, P., Mendelsohn, M., Mann, J.J., Brunner, D., and Hen, R. (1998).

Serotonin receptor 1A knockout: an animal model of anxiety-related disorder. Proc Natl Acad Sci U S A 95, 14476-14481.

Ratner, C., Ettrup, A., Bueter, M., Haahr, M.E., Compan, V., le Roux, C.W., Levin, B., Hansen, H.H., and Knudsen, G.M.

(2012). Cerebral markers of the serotonergic system in rat models of obesity and after Roux-en-Y gastric bypass. Obesity (Silver Spring) 20, 2133-2141.

Richardson-Jones, J.W., Craige, C.P., Guiard, B.P., Stephen, A., Metzger, K.L., Kung, H.F., Gardier, A.M., Dranovsky, A., David, D.J., Beck, S.G., et al. (2010). 5-HT1A autoreceptor levels determine vulnerability to stress and response to antidepressants. Neuron 65, 40-52.

Salmon, P. (2001). Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clin Psychol Rev 21, 33-61.

Walsh, B.T. (2013). The enigmatic persistence of anorexia nervosa. Am J Psychiatry 170, 477-484.

West, G.B., Brown, J.H., and Enquist, B.J. (1997). A general model for the origin of allometric scaling laws in biology.

Science 276, 122-126.

Yadav, V.K., Oury, F., Suda, N., Liu, Z.W., Gao, X.B., Confavreux, C., Klemenhagen, K.C., Tanaka, K.F., Gingrich, J.A., Guo, X.E., et al. (2009). A serotonin-dependent mechanism explains the leptin regulation of bone mass, appetite, and energy expenditure. Cell 138, 976-989.

SUPPLEMENTAL INFORMATION

Supplemental information includes Results, Experimental Procedures, Statistics, six figures, one table and video.

LEGENDS

Figure 1. The mPFC-5-HT

R Exert a Negative Control of Food Intake in the Basal Condition (A-C) Daily food intake, body weight, food intake / active tissue (body weight

)

(D) Food intake / active tissue 3 h post-treatment

(E, F) mPFC-5-HT

R mRNA 5 and 77 h respective post-infusion in the mPFC of si5-HT

R in 129SvPas WT and HSV5-HT

R in 129SvTer mice of both genotypes

(G) 5-HT

R binding sites (OD: optical density evaluated on rectangular surface 80 µm

)

(H) Labeling of the [

I]SB207710 (5-HT

R antagonist) visualized from autoradiographs in frontal brain sections.

$$$

p < 0.0001 differences between 129SvPas and 129SvTer and

p < 0.001, genotypes;

p < 0.05,

p < 0.01,

p <

0.001 compared with siCt and

p < 0.05,

p < 0.001 with HSVLacZ;

p < 0.05 compared to siCt-WT and,

p <

0.05 to si5-HT

R-WT following ANOVA: Two-way repeated measured (A-C) n = 30-60 F(1,88) = 88.27 p <

0.0001; One-way (D) n = 10-16 F(3,71) = 12.32 p < 0.0001, (E) n = 6-9 F(1,13) = 4.81 p < 0.05; Two-way (F) n =

4-5 HSV5-HT

R in 129SvTer mice of both genotypes, treatment F(1,14) = 40.11 p = 0.0001, (G) n = 10-20

genotype x treatment F(5,88) = 86.1 p < 0.0001. n = number of mice/group.

Figure 2. Causal Relationship Between the 5-HT

Rs in the mPFC and Hypophagia Due to Stress (A-D) Daily food intake for 21 h

(E) Experimental design

(F-H) mPFC-5-HT

R mRNA immediately after stress

$

p < 0.05,

p < 0.0001 and

p < 0.05,

p < 0.0001 respective differences between 129SvPas and 129SvTer and, genotypes;

p < 0.05,

p < 0.01 compared with siCt and

p < 0.05,

p < 0.01,

p < 0.001 with HSVLacZ;

p <

0.05,

p < 0.01 compared to unstressed following ANOVA: Two-way repeated measured (A) n = 14-17 treatment F(1,58) = 4.51 p < 0.05, time F(4,232) = 12.74 p < 0.0001, time x treatment F(4,232) = 6.08 p < 0.05 and time x genotype x treatment F(4,232) = 2.30 p < 0.05; (B-C) n = 5-10 time F(4,168) = 11.6 p < 0.0001, time x stress F(4,168) = 7.07 p < 0.0001, time x stress x genotype F(4,168) = 2.5 p < 0.05, time x stress x treatment F(4,168) = 2.6 p< 0.05, (D) n = 13-16 time x stress x treatment F(4, 224) = 3 p < 0.05, stress F(4, 224) = 6.3 p < 0.0001, time F(4, 224) = 16.9 p < 0.0001, treatment F(4, 224) = 2.7 p < 0.05 and (D) additional two-way repeated measures revealed difference in basal food intake between 129SvPas and 129SvTer WT mice F(1,80) = 45 p < 0.0001; Two- way (F-G) n = 3-5 treatment F(1,28) = 15.4 p < 0.001; (H) n = 6-7 treatment F(1,23) = 25.6 p < 0.0001, stress F(1,23) = 5.3 p < 0.0001, stress x treatment F(1,23) = 3.1 p = 0.05; One-way (F) stress in HSVLacZ-transduced WT mice F(1,8) = 8.8 p < 0.05, (B-C) one-way ANOVA day 6 stress F(1,40) = 12.5 p < 0.01, genotype x treatment x stress F(1,40) = 4.0 p < 0.05. n = number of mice/group.

See also Figure S2.

Figure 3. Stress-induced Changes in the 5-HT Parameters in the DR depends on the mPFC-5-HT

Rs (A) Fos mRNA in the DR/MR immediately after stress

(B) Extracellular 5-HT in the DR. Data corresponding to “HSVLacZ- and HSV5-HT

R-transduced WT and, HSVLacZ- and HSVLacZ-transduced KO 5-HT

R + stress” are graphically grouped behind the green circles (C) Summary: x absent, é increases, éé, high increases, ê decreases, = unchanged

(D, E) Respective levels of SERT and 5-HT

R mRNA in the DR/MR immediately after stress (F) 5-HT

R binding sites at the DR/MR level

&

p < 0.05,

p < 0.01,

p < 0.001 stress effect;

p < 0.05,

p < 0.001 differences between genotypes and;

p <

0.05,

p < 0.01 treatments following two-way ANOVA: (A) genotype x stress F(1,35) = 4.3 p < 0.05, treatment x stress F(1,35) = 5.4 p < 0.05, genotype F(1,35) = 5.1 p < 0.05, stress F(1,35) = 27.0 p < 0.0001; (B) genotype x stress x treatment x time F(1,29) = 3.2 p < 0.01, genotype x stress x treatment F(1,29) = 6.8 p < 0.05, stress F(1,29)

= 16.2 p < 0.0004; (D) genotype x stress x treatment F(1,31) = 4.2 p < 0.05, genotype x stress F(1,31) = 3.9 p <

0.05, genotype x treatment F(1,31) = 7.3 p < 0.01, genotype F(1,31) = 64.2 p = 0.0001, stress F(1,31) = 21.9 p = 0.0001, treatment F(1,31) = 14.6 p = 0.0006; (F) genotype x stress x treatment F(1,13) = 9.9 p < 0.01, stress x treatment F(1,13) = 5.9 p < 0.05, genotype F(1,13) = 11.1 p < 0.01, stress F(1,13) = 32.6 p < 0.0001, treatment F(1,13) = 14.9 p < 0.01. n = 5-7 mice/group.

See also Figure S3 and Table S1.

Figure 4. Early Anorexia in Stressed Mice with mPFC-5-HT

R Overexpression and DR-5-HT

R Blockade

(A, B) Daily food intake for 21 h

(C, D) In identical mice, body weight gain or loss (body weight evaluated each day minus the first day of the test)

(E) Experimental design.

&&

p < 0.01,

p < 0.001 stress effect,

p <

0.05,

p < 0.01 differences between DR treatments in stressed mice with mPFC-HSV5- HT

Rs following two-way repeated measures ANOVA: (A, B) stress x food intake F(1,51) = 7.0 p < 0.01, time F(4,204) = 8.3 p < 0.0001, effect of DR treatment over time F(4,204) = 2.6 p < 0.05. ANOVA analyses for each day:

stress, day 2 F(1,51) = 40.7 p < 0.0001, day 3 F(1,51) = 14.0 p < 0.001] and day 7, mPFC F(1,51) = 4.8 p < 0.05 and DR F(1,51) = 7.2 p

< 0.01) treatment effect on food intake; (C, D) stress F(1,51) = 7.0 p < 0.01, time F(4,204) = 4.7 p < 0.01 and mPFC x DR treatment F(1,51)

= 3.8 p < 0.05 . ANOVA analyses for each day:

stress, day 2 F(1,51) = 8.5 p < 0.01, day 3 F(1,51) = 8.1 p < 0.001, day 4 F(1,51) = 5.5 p

< 0.05] and day 7 mPFC x DR treatment F(1,51) = 5.1 p < 0.05. n = 6-10 mice/group.

See also Figure S4.

SUPPLEMENTAL EXPERIMENTAL PROCEDURES

Animals

The generation of 5-HT

R KO and WT mice on a 129SvTer background from heterozygous breeding has been described (Compan et al., 2004). The experiments involving siRNA and pharmacological treatments required the utilization of WT animals only. However, 129SvTer WT mice cannot be externally purchased and the local rules of Functional Genomic facilities (Montpellier, France) exclude the breeding of WT mice. The 129SvTer WT mice can therefore only be obtained from the breeding of 129SvTer heterozygote mice but were utilized with mutant mice at an identical age (Compan et al., 2004). We exclude to sacrifice unnecessarily extra KO mice. Therefore, WT mice from a 129SvPas similar genetic background were purchased (Charles River) in order to conduct the experiments where KO mice were not required. Mice of both genotypes and 129/SvPas WT mice were housed as same-sex littermates (n = 5 per cage) with food and water available ad libitum in a temperature - controlled environment with a 12-h light/dark cycle (light onset at 07:00).

Surgery As we described (Jean et al., 2007; Jean et al., 2012), mice were anesthetized by i.p. injection of ketamine (60 mg/kg)

and xylazine (15 mg/kg) and placed in a David Kopf stereotaxic frame. Each compound used was dissolved in NaCl (9‰) and stored at 4°C until use. A sterile 26-gauge stainless guide cannula was unilaterally implanted and fixed with dental cement 48 h before the onset of the habituation period of any experiments, in the left mPFC, and / or DR or left NAc, at the following respective coordinates from the bregma: A + 1.8 mm, H - 2.0 mm, L + 0.2 mm; A - 4.7 mm, H - 3.2 mm, L + 0.0 mm; A + 1.6 mm, H - 4.3 mm, L + 0.7 mm (Franklin and Paxinos, 1997). Each compound was infused, through a stainless cannula, which was inserted in the guide cannula, in the mPFC and / or the DR or, in the NAc (1 µl/min) and was connected with a micro- catheter to a microsyringe nanopump (CE, Myneurolab, St Louis, MO) in freely moving mice after the habituation period. The localization of the injection site was assessed for each mouse.

Radioimmunoassay As we described (Compan et al., 2004), blood samples from naïve stressed and unstressed mice, single

housed for 7 days, were collected after a small tail incision immediately after a stress period of 110 min. Identical blood samples were used to evaluate the levels of glucose. Blood samples were immediately centrifuged (4°C), and plasma samples were stored in -80°C until analyses. The levels of corticosterone were evaluated using appropriate [

I]- radioimmunoassay kit (ICN Biomedicals, Costa Mesa, CA).

Nucleic Acid Treatments As we conducted (Jean et al., 2007; Jean et al., 2012), series of 129SvPas WT mice received, in the

mPFC or NAc, an acute injection of 1 µl containing either NaCl (9‰) or double-stranded si5-HT

R (0.05 µg/µl, mouse mRNA sense, 5’- UAAGUCUUUCAGACGUGCC99-3’; antisense, 5’-GGCACGUCUGAAAGACUUA99-3’) or control (siCt: sense, 5’-AUGAUGGCAACUGAUCGAC99-3’; antisense, 5’-GUCGAUCAGUUGCCAUCAU99-3’ (0.05 µg/µl, si5-HT

R control (Eurogentec) contained identical nucleotides but in random order and had no significant homology with any known rodent mRNAs. Generation of HSV-5-HT

R construct was described (Lucas et al., 2005). A cDNA fragment comprising the open reading frame of the Htr4 gene encoding 5-HT

R (obtained from the full-length 5-HT

splice variant) was inserted in the HSVPrpUC amplicon vector. The amplicon construct was packaged with the 5dl1.2 helper virus, purified on a sucrose gradient and suspended in 10% sucrose. The HSV immediate-early gene promoter IE4/5 regulates the transgene expression. The HSV plasmids were stored at -80°C. The HSV-LacZ construct and HSV5-HT

R were injected, in the mPFC, at the concentration of 10

infectious units/ml.

Quantitative Real Time PCR As we described (Conductier et al., 2006; Jean et al., 2007; Jean et al., 2012), the NAc and mPFC

were micro-dissected (1.2 mm

) from 1 mm-thick section [NAc: A + 1.6 mm, mPFC: A + 1.8 mm from the bregma, following the landmarks of the stereotaxic atlas (Franklin and Paxinos, 1997)] to treat total mRNA and treat cDNA in reactions containing 5- HT

R primers (sense and antisense primers that hybridize, respectively, to nucleotides 688-705 and 631-654 of the mRNA sequence of mouse 5-HT

R: NM-008313). Total mRNA was isolated, treated with DNAse, and reverse transcribed.

First-strand cDNA was used as a template for QR-PCR amplification (ABI Prism 7000, Applied Biosystems) in 10 µl reactions containing 300 nM of each of the 5-HT

R mRNA primers and a master mix (SybR Green) including Taq DNA polymerase.

Samples were successively incubated (50°C, 2 min; 95°C, 10 min), followed by 40 cycles of denaturation (95°C, 15 s), annealing and extension (60°C, 1 min). The size of fragments was assessed using electrophoretic separation in nusive 3:1 agarose gel. The relative expression level of the Htr4 gene was evaluated by QR-PCR of two housekeeping genes (aldolase 3, 18S rRNA). Their expression was not significantly different between genotypes or treatments.

Biochemical Analyses As we described (Dusticier and Nieoullon, 1987), the levels of tissue 5-HT and 5-HIAA were evaluated

in brain tissue samples of stressed and unstressed WT and 5-HT

R KO mice sacrificed at the end of the stress period (110 min). The brains were removed and kept on ice. Tissue samples containing hypothalamus, NAc, striatum, amygdala, dorsal hippocampus and DR/MR were then micro-dissected from 1 mm thick sections at -20°C using micro-punches following the landmarks of the stereotaxic atlas [Hypothalamus -0.46 mm, NAc +1.6 mm, striatum +1.0 mm, amygdala -3.2 mm, dorsal hippocampus -2.2 mm, DR / MR -4.7 mm from the bregma (Franklin, 1997)]. All samples were weighed and stored at -80 °C.

As we reported (Doly et al., 2008), the levels of extracellular 5-HT were evaluated in dialysates collected by utilizing dialysis

probes that are equipped with a Cuprophane membrane (length 1 mm, diameter 0.24 mm, cut-off 6000 Da, CMA

Microdialysis, Solna, Sweden). Probes were perfused at a constant rate of 1 µl / min with artificial CSF containing 145 mM

NaCl, 2.7 mM KCl, 1.0 mM MgCl2, 1.2 mM CaCl2 and 2.0 mM de Na2HPO4 adjusted to pH 7.4. Dialysates were collected

every 20 min, 240 min after the beginning of perfusion, by which time a steady state was achieved (2 h). WT and 5-HT

R KO

mice were treated with HSVLacZ or HSV5-HT

R 3 days and 3 h before the onset of the restraint stress period. Dialysate

samples (10 µl) were treated without any purification into an HPLC system, as described (Doly et al., 2008). Absolute basal 5-

HT levels in dialysate collected from the DR were taken as the mean ± SEM of 4-5 values.

Receptor Autoradiography As we described (Compan et al., 1996; Compan et al., 1998; Compan et al., 2004; Conductier et al.,

2006; Jean et al., 2007), frozen serial coronal brain 12 µm thick sections from treated and control mice (si5-HT

R, siCt, HSV5- HT

R, HSVLacZ) were processed for ligand-binding autoradiography of 5-HT

Rs with the specific antagonist [

I]SB207710.

Briefly, sections were incubated in the appropriate buffer supplemented with 10 µm pargyline, 0.01% ascorbic acid and [

I]SB207710 (specific activity, 2000 Ci/mmol; final concentration, 0.02 nM; Amersham, Piscataway, NJ), at 37°C for 30 min. Non-specific binding was determined on consecutive sections incubated in the presence of 1 µM GR113808 [Sigma- Aldrich, St. Louis, MO]. Additional frozen serial coronal brain 12 µm thick sections from treated (HSV5- HT

R) and control (HSVLacZ) stressed and unstressed WT and 5-HT

R KO mice were processed for ligand binding autoradiography of 5-HT

R with [

H]8-hydroxy-2-di-n-propylamino-tetralin ([

H]8-OH-DPAT). Briefly, sections were incubated in Kreb’s solution [30 mM Tris, 118 mM NaCl, 1.2 mM CaCl

, 1.2 mM MgCl

, 4.8 mM KCl (pH 7.4), 2 x 40 min at 4°C] for [

H]8-OH-DPAT binding. Tissue was then incubated at room temperature (22°C) for 1 h in the appropriate buffer supplemented with 10 µM pargyline, 0.01% ascorbic acid and [

H]8-OH-DPAT (Amersham, UK; specific activity, 129.5 Ci ⁄ mmol; final concentration 2 nM). Non-specific binding was determined on consecutive sections incubated in the presence of 10 µM 5-HT (Sigma-Aldrich, France) for [

H]8-OH-DPAT. All sections were exposed to KODAK X-Omat Blue (PerkinElmer, Waltham, MA) for 5 days at 4 °C (19) for [

I]SB207710 and to [

H]hyperfilm and processed in a Kodak D-19 developer following an exposure of 2 months at 4°C for [

H]8-OHDPAT. Sections from both groups were processed together in order to compare signals on the same film.

Feeding Test As we described (Compan et al., 2004; Jean et al., 2007), 48 h after surgery, mice were isolated in individual

cages for a baseline period with ad libitum access to food (pellet form, 16.5% crude proteins, 3.6% crude fat, 4.6% crude fiber, 5.2% ash) and tap water. Mice were fed with water given ad libitum (9:00 to 9:00). Water intake was not evaluated. The restraint stress, i.e. forced immobilization was applied in a ventilated 50 ml polypropylene tubes for 110 min (Compan et al., 2004). Each experiment was divided into three periods: baseline (5 days), the day (Day 5) of treatments and restraint stress, and the recovery period (3 days). The day of treatment includes an initial handling for weighing (t = 0 min), the injection, at t = 10 min, of 1 µl solution in the mPFC (1 µl/min, microsyringe nanopump, CE; myNeuroLab, St. Louis, MO), containing double-stranded si5-HT

Rs (50 ng/µl), si5-HT

Rs control (siCt, 50 ng/µl), BIMU8 (40 ng/µl, Boehringer Ingelheim, Ingelheim, Germany), RS39604 (40 ng/µl, Tocris, Ellisville, MO, USA) or NaCl (9‰) and, the onset of stress at t = 3 h for a duration of 110 min (the duration from t = 0 min to the end of stress is 5 h). The injection of HSV5-HT

R, HSVLacZ (10

infectious units/ml) in the mPFC was performed 3 days and 3 h before stress (Days 0-2: Figures 1-3) and, was or not combined with local injection in the DR of WAY100635 (45 ng/µl, 1 µl/min) the 6

day (i.e. 19 h after the end of the stress period). The injection of MK801 (0.5 ng/µl, Tocris, Ellisville, MO, USA) or NaCl (9‰) in the DR has been performed immediately after stress.

Additional mice were sacrificed either 5 h (Day 5) or, 3 days and 5 h (77 h, immediately after stress, Day 5) following respective post-infusion of siRNA and HSV for quantitative RT-PCR and/or autoradiography analyses. Food intake of isolated mice was evaluated either daily (at 09:00, 2 h after the start of light cycle) following treatments or, 3 h in additional isolated mice in the basal condition following injection of treatments in the mPFC (si5-HT

Rs, siCt, HSV5-HT

R, HSVLacZ, BIMU8, RS39604, NaCl). In each experiment, mice were divided into unrestrained and restrained groups matched for average weight.

On the day of stress, unrestrained and restrained mice were identically food and water deprived for 110 min. An interval of 3 min was maintained between each mouse in order to apply stress. Following 110 min of a stress period, restrained and unrestrained animals returned to their regular cages with free access to food and water.

Open Field As we previously described (Compan et al., 2004), mice were tested for 30 min, 1 h after the end of the restraint

stress period of 110 min. The open field test environment is a square chamber with an inside area that measures 43.2 x 43.2 x 30.5 cm. Mice were placed in the center and monitored with 32 infrared light sources spaced 0.5 inches apart (1.25 cm) (Med Associates, Georgia, VT) specifically adapted to record the location and the traveled path length.

SUPPLEMENTAL REFERENCES

Amargos-Bosch, M., Bortolozzi, A., Puig, M.V., Serrats, J., Adell, A., Celada, P., Toth, M., Mengod, G., and Artigas, F.

(2004). Co-expression and in vivo interaction of serotonin1A and serotonin2A receptors in pyramidal neurons of prefrontal cortex. Cereb Cortex 14, 281-299.

Bortolozzi, A., Amargos-Bosch, M., Toth, M., Artigas, F., and Adell, A. (2004). In vivo efflux of serotonin in the dorsal raphe nucleus of 5-HT1A receptor knockout mice. J Neurochem 88, 1373-1379.

Czyrak, A., Czepiel, K., Mackowiak, M., Chocyk, A., and Wedzony, K. (2003). Serotonin 5-HT1A receptors might control the output of cortical glutamatergic neurons in rat cingulate cortex. Brain Res 989, 42-51.

Franklin, K.B.J., and Paxinos, G. (1997). The mouse brain in stereotaxic coodinates. (San Diego: Academic press).

Haj-Dahmane, S., Hamon, M., and Lanfumey, L. (1991). K+ channel and 5-hydroxytryptamine1A autoreceptor interactions in the rat dorsal raphe nucleus: an in vitro electrophysiological study. Neuroscience 41, 495-505.

Hajos, M., Richards, C.D., Szekely, A.D., and Sharp, T. (1998). An electrophysiological and neuroanatomical study of the medial prefrontal cortical projection to the midbrain raphe nuclei in the rat. Neuroscience 87, 95-108.

Lee, H.S., Kim, M.A., Valentino, R.J., and Waterhouse, B.D. (2003). Glutamatergic afferent projections to the dorsal raphe nucleus of the rat. Brain Res 963, 57-71.

Lucas, G., Compan, V., Charnay, Y., Neve, R.L., Nestler, E.J., Bockaert, J., Barrot, M., and Debonnel, G. (2005).

Frontocortical 5-HT4 receptors exert positive feedback on serotonergic activity: viral transfections, subacute and chronic

treatments with 5-HT4 agonists. Biol Psychiatry 57, 918-925.

Lucas, G., and Debonnel, G. (2002). 5-HT4 receptors exert a frequency-related facilitatory control on dorsal raphe nucleus 5- HT neuronal activity. Eur J Neurosci 16, 817-822.

Miquel, M.C., Doucet, E., Riad, M., Adrien, J., Verge, D., and Hamon, M. (1992). Effect of the selective lesion of serotoninergic neurons on the regional distribution of 5-HT1A receptor mRNA in the rat brain. Brain Res Mol Brain Res 14, 357-362.

Penas-Cazorla, R., and Vilaro, M.T. (2014). Serotonin 5-HT receptors and forebrain cholinergic system: receptor expression in identified cell populations. Brain structure & function.

Peyron, C., Petit, J.M., Rampon, C., Jouvet, M., and Luppi, P.H. (1998). Forebrain afferents to the rat dorsalraphe nucleus demonstrated by retrograde and anterograde tracing methods. Neuroscience 82, 443-468.

Pompeiano, M., Palacios, J.M., and Mengod, G. (1992). Distribution and cellular localization of mRNA coding for 5-HT1A receptor in the rat brain: correlation with receptor binding. J Neurosci 12, 440-453.

Santana, N., Bortolozzi, A., Serrats, J., Mengod, G., and Artigas, F. (2004). Expression of serotonin1A and serotonin2A receptors in pyramidal and GABAergic neurons of the rat prefrontal cortex. Cereb Cortex 14, 1100-1109.

Seguela, P., Watkins, K.C., and Descarries, L. (1989). Ultrastructural relationships of serotonin axon terminals in the cerebral cortex of the adult rat. J Comp Neurol 289, 129-142.

Sesack, S.R., Deutch, A.Y., Roth, R.H., and Bunney, B.S. (1989). Topographical organization of the efferent projections of the medial prefrontal cortex in the rat: an anterograde tract-tracing study with Phaseolus vulgaris leucoagglutinin. J Comp Neurol 290, 213-242.

Smiley, J.F., and Goldman-Rakic, P.S. (1996). Serotonergic axons in monkey prefrontal cerebral cortex synapse predominantly on interneurons as demonstrated by serial section electron microscopy. J Comp Neurol 367, 431-443.

Sprouse, J.S., and Aghajanian, G.K. (1987). Electrophysiological responses of serotoninergic dorsal raphe neurons to 5-HT1A and 5-HT1B agonists. Synapse 1, 3-9.

Steinbusch, H.W. (1981). Distribution of serotonin-immunoreactivity in the central nervous system of the ratcell bodies and terminals. Neuroscience 6, 557-618.

Van Bockstaele, E.J., Biswas, A., and Pickel, V.M. (1993). Topography of serotonin neurons in the dorsal raphe nucleus that send axon collaterals to the rat prefrontal cortex and nucleus accumbens. Brain Res 624, 188-198.

Vertes, R.P. (2004). Differential projections of the infralimbic and prelimbic cortex in the rat. Synapse 51, 32- 58.

Wang, Q.P., Ochiai, H., and Nakai, Y. (1992). GABAergic innervation of serotonergic neurons in the dorsal raphe nucleus of

the rat studied by electron microscopy double immunostaining. Brain Res Bull 29, 943-948.