Changes in food intake and anxiety-like behaviors after clonidine injected into the median raphe nucleus

Behavioural Brain Research 212 (2010) 71–77

Contents lists available at ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Changes in food intake and anxiety-like behaviors after clonidine injected into the median raphe nucleus Samira Schultz Mansur ∗ , Mariana Graciela Terenzi, José Marino Neto, Moacir Serralvo Faria, Marta Aparecida Paschoalini Department of Physiological Sciences, Centre of Biological Sciences - CCB, Federal University of Santa Catarina (UFSC), 88040-970, Florianópolis, SC, Brazil

a r t i c l e

i n f o

Article history: Received 12 October 2009 Received in revised form 19 March 2010 Accepted 23 March 2010 Available online 30 March 2010 Keywords: Median raphe nucleus ␣2 -adrenergic receptor clonidine Clonidine Food intake Anxiety-like behavior Rat

a b s t r a c t Serotonergic neurons in the median raphe nucleus (MnR) are stimulated by ␣1 -adrenergic agonists and inhibited by ␣2 agonists. This study investigated the participation of MnR ␣2 -adrenergic receptors in the control of anxiety-like behavior and feeding as an attempt to establish a functional association between these behaviors. The ␣2 -adrenergic agonist clonidine (CLON) was injected into the MnR (0, 0.2, 2, 6, 20 nmol), into the pontine nucleus (Pn) or into the mesencephalic reticular formation (mRt) (0.2, 20 nmol) of free-feeding rats. The animals were exposed to the elevated plus-maze to evaluate spatial-temporal and ethological variables. Subsequently, the ingestive and non-ingestive behaviors were recorded during 30 min and the quantity of food and water consumed were measured. The lowest dose of CLON injected into the MnR decreased the total risk assessment (TRA) frequency, an ethological parameter of anxiolyticlike effect, but did not change feeding behavior. The highest dose of CLON injected into the MnR increased the TRA frequency, an anxiogenic-like effect. Similar result was observed after CLON injected into the Pn and mRt at the highest dose. In addition, clonidine at the highest dose caused hyperphagy accompanied by a reduction in the latency to start eating and an increase in feeding frequency when injected into the MnR but not in the Pn or mRt. These data indicate that MnR ␣2 -adrenergic receptors participate in the control of anxiety-like and feeding behaviors, probably decreasing the facilitatory influence on MnR serotonergic neurons. The present results suggest that these behaviors involve independent neural pathways. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The raphe nuclei are a collection of neurons with distinct morphologies, projections and neurochemical characteristics. They flank the midline along the rostrocaudal extension of the brainstem and constitute the main source of serotonin (5-hydroxytryptamine, 5-HT) in the central nervous system [2,50,69,74]. One of the major sites of serotonin projections in the brain is originated in neurons located in the median raphe nucleus (MnR) [37,50]. These projections ascend within the medial forebrain bundle and mainly target structures of the forebrain [10,11] implicated in the regulation of behaviors like anxiety and food intake [20,35,37]. The MnR is one of the brain areas with the highest density in 5-HT1A receptors [2,15,38,40,41,67]. The great majority of these receptors is found on serotonergic neurons and function as autoreceptors that regulate the synthesis and release of 5-HT in their projection areas [2,31]. Injections of the 5-HT1A receptor agonist 8-

∗ Corresponding author. Tel.: +55 48 37219352; fax: +55 48 37219672. E-mail address: [email protected] (S.S. Mansur). 0166-4328/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2010.03.046

hidroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT) into the MnR inhibit 5-HT cell firing and reduce its release in both the cell body and terminals [2,5,6,11,30,65]. Injections of 8-OH-DPAT into the MnR have been reported to increase food intake [16]. It has also been reported that the activation of presynaptic 5-HT1A autoreceptors located in the MnR decreases anxiety-like behavior in rats exposed to the plus-maze [3,12,27,51]. This anxiolytic effect is antagonized by MnR infusion of the 5-HT1A receptor antagonist WAY 100635 [27]. Furthermore, intra-MnR injection of the GABAA receptor agonist muscimol inhibited the firing of 5-HT neurons and may be involved in anxiolytic-like effects [38]. In addition to the high density of 5-HT receptors, a great density of ␣1 and ␣2 -adrenergic receptors is also found in the MnR. It has been reported that endogenous noradrenaline in the MnR exerts facilitatory control of 5-HT release via ␣1 -adrenergic receptors and inhibitory control via the ␣2 subtype [1]. MnR neurons receive noradrenergic innervation from the locus coeruleus/subcoeruleus, lateral tegmental area and A1 and A2 cell groups [2,15,36]. The ␣2 -adrenergic agonist clonidine (CLON) injected into the MnR decreases the dialysate level of 5-HT to 30% of basal values into this nucleus, while ␣2 -adrenergic antagonists into the MnR enhance its

72

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77

release [1]. In addition, injections of ␣1 -adrenergic receptor antagonists into the MnR decrease extracellular levels of 5-HT in this nuclei as well as in forebrain structures such as the hippocampus and the striatum [1,11]. Previous studies in our laboratory revealed that adrenaline (AD) but not noradrenaline injected into the MnR decreased food intake and shortened meal duration in food restricted rats [49]. On the other hand, injection of AD into the MnR of free-feeding rats increased food intake, feeding frequency and decreased latency to start feeding [22]. In view that inhibitory action of the 5-HT on feeding behavior has been extensively reported [17,19,23,64,79] and that the serotonergic innervation of the hypothalamic paraventricular nucleus, an important region where 5-HT has been implicated in feeding regulation, arises from MnR serotonergic projections [23,24,48] we suggested that in food restricted animals, the ADinduced hypophagia may be due to the activation of ␣1 -adrenergic receptors on 5-HT MnR neurons resulting in 5-HT release and an indirect inhibitory action of AD on feeding [49]. In free-feeding rats, the food intake effects evoked by AD injections may be attributed to the activation of ␣2 -adrenergic receptors on 5-HT MnR neurons that could act to suppress the release of 5-HT and its inhibitory action on feeding behavior [22]. Furthermore, there were similarity between feeding effects elicited by AD and 8-OH-DPAT. It has been reported that food intake evoked by 8-OH-DPAT treatment vary depending on nutritional state. Hypophagia in fasted and hyperphagia in non-food deprived-animals have been seen after treatment with 8-OH-DPAT [24,48,64]. The hyperphagia evoked by 8-OH-DPAT was attributed to activation of 5-HT1A somatodentritic autoreceptor found in the MnR, leading to an inhibition of 5-HT neuronal activity and reduced synthesis and release of 5-HT in forebrain terminal areas [17,23]. This evidence reinforces the suggestion that MnR serotonergic neurons seem to mediate the adrenergic effects on feeding behavior. The aim of this study is to test the suggestion that the feeding effects evoked by AD injection into the MnR of free-feeding rats are due to the activation of ␣2 -adrenergic receptor and to establish a relationship between food intake and anxiety-like behavior. 2. Methods and materials 2.1. Animals Male Wistar rats (weight 260–290 g at the time of surgery) were group-housed in a room at 22–24 ◦ C, 12:12 light–dark cycle (lights on at 6:00 AM, provided by a 100-W bulb 2.75 m above the center of the room, approximately 100 l× at central floor level) with standard rodent chow and water available ad libitum. The animals were housed in groups of five per cage until the onset of the experiments. After surgery, rats were housed individually. The experimental procedures were conducted in compliance with the recommendations of the Ethics Commission for the use of Experimental Animals of the Federal University of Santa Catarina, SC, Brazil. All efforts were made to minimize the number of animals used and their suffering. 2.2. Stereotaxic surgery

volumes (0.2 ␮l) were administered over a period of 60 s and a further 60 s was allowed for the solution to diffuse from the needle. Clonidine hydrochloride (CLON), an ␣2 -adrenergic agonist, used at the doses of 0.2, 2, 6 and 20 nmol, was purchased from SIGMA (USA) and dissolved in 0.9% sterile saline. Control groups were injected with vehicle alone. The behavioral active doses of CLON were based on those used in previous report [43]. 2.4. Experimental procedures Experiment 1 was designed to evaluate the effects of CLON injections into the MnR on anxiety-like, ingestive and non-ingestive behaviors. During the experimental session, naïve free-feeding rats were treated in the MnR with vehicle (saline, n = 6) or CLON at doses of 0.2 nmol (n = 6), 2 nmol (n = 6), 6 nmol (n = 6) or 20 nmol (n = 8). After the injection, the animals were placed in the elevated plus-maze (EPM) to record the spatial-temporal and ethological parameters related to anxiety. In sequence, the rats were transferred to the feeding recording chamber to evaluate the ingestive and non-ingestive-behaviors. In experiment 2, the effective doses detected in experiment 1 were injected into areas close to the MnR, such as the pontine nuclei (Pn) and mesencephalic reticular formation (mRt) and then the same behaviors of experiment 1 were analyzed to confirm whether the CLON effects on anxiety-like and feeding behaviors were site-specific. Free-feeding naïve rats were treated in the Pn with vehicle (saline, n = 8) or CLON at doses of 0.2 nmol (n = 8) or 20 nmol (n = 14) and in the mRt with vehicle (saline, n = 4) or CLON at doses of 0.2 nmol (n = 8) or 20 nmol (n = 9). Each rat received only one microinjection. 2.4.1. Anxiety-like behavior Anxiety-like behavior was monitored during 5 min using the EPM. Each experimental session was recorded by a webcam and subsequently coded using the Etholog 2.25 software [56]. During the session, the conventional spatial-temporal measures such as the total number of entries into either the open or the enclosed arms, the percentage of entries and the time spent inside the open arm and the ethologically derived variables like the frequencies of total risk assessment (TRA), total head dipping, rearing and grooming were measured. The descriptions of the ethological variables are as follows: risk assessment: investigation of the surroundings with the forepaws and head; head dipping: protruding the head over the edge of an open arm and scrutinizing in any direction; rearing: rising on the hind limbs; grooming: cleaning any part of the body surface with the tongue, teeth and/or forepaws [66]. The frequencies of risk assessment and head dipping recorded in this study were the “total” measured in protected and unprotected areas of the EPM. 2.4.2. Food intake behavior Immediately after the exposure to the EPM, each animal was placed for 30 min in a food cage containing a known weight of standard rodent chow and a previously measured volume of tap water. At the end of the session, the difference between food and water at the beginning and at the end was taken as the amount of food or water consumed. The experiment was recorded by a webcam for subsequent behavioral analysis with the Etholog 2.25 [56]. During the session the latencies to start feeding and drinking, durations and frequencies of feeding and drinking as well as the frequencies of locomotion, non-locomotion, immobility, rearing and grooming were evaluated. See Lopes et al. for details about these procedures [47]. 2.5. Histological analysis At the end of the experiments, the animals were deeply anaesthetized and transcardially perfused with saline (0.9%) and formalin (10%). The brains were removed, maintained in formalin and sliced (vibratome) in the coronal plane (100 ␮m). Sections were mounted on gelatinized slides and stained with cresyl violet. The cannulae placements were identified under a microscope by comparison of the sections with the photographs and diagrams of the atlas of The Rat Brain [58]. Only data from rats with cannulae correctly placed into the MnR, Pn or mRt were included in the study.

The rats were anesthetized (ip) with ketamine hydrochloride (87 mg kg−1 ) and xylazine (13 mg kg−1 ) mixture and stereotaxically implanted with a unilateral stainless steel guide cannula (30 G, 18 mm length). The target to reach was a location 2 mm above the MnR based on the atlas of The Rat Brain [58]. The following coordinates from bregma were used: AP = −7.8 mm; L = 3.0 mm; DV = 7.0 mm from the skull surface, at an angle of 20◦ from the vertical plane to avoid the sagital sinus and the cerebral aqueduct. The same surgical procedures and AP and L coordinates were adopted for the Pn and mRt, except the DV measures that had a difference of 2 mm above or below the MnR for the mRt and Pn, respectively [58]. The cannula was anchored to the skull with dental cement and the whole implant stabilized with jeweller screws and more dental cement. A removable stylet was introduced to keep the cannula free from blockages until the day of the experiment.

2.6. Statistical analysis

2.3. Drug injections

3. Results

Injections were made using an inner cannula (33 G, 20 mm length) extending 2 mm beyond the ventral tip of the guide cannulae and connected by polyethylene tubing to a Hamilton microsyringe (1 ␮l) fitted to an injection pump. The injected

Fig. 1 shows sequential coronal sections of the injection sites within the MnR, Pn and mRt. Histological analysis indicated that 32

In experiment 1, the effects of saline (0) and CLON (0.2, 2, 6, 20 nmol) into the MnR on anxiety-like, ingestive and non-ingestive behaviors were analyzed by oneway analysis of variance (ANOVA). In experiment 2, the effects of treatment with saline (0) or CLON (0.2 and 20 nmol) in the MnR, Pn and mRt on anxiety-like behavior were analyzed by two-way ANOVA (drug-dose × injected area). The same statistical procedure was used to analyze the feeding effects evoked by injections of saline (0) and CLON (20 nmol) into these structures. Results are expressed as mean ± standard error of mean (S.E.M.). When appropriate, the ANOVA was followed by Duncan’s post hoc test for multiple comparisons. Only probability values less than 5% were considered significant.

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77

73

Fig. 1. Photomicrographs and schematic drawings (adapted from Paxinos and Watson [58]) of coronal sections through the mesencephalic reticular formation (mRt), median raphe nucleus (MnR) and pontine nucleus (Pn), illustrating the approximate sites (䊉) of clonidine injections into the MnR (6 animals/treatment, except for 20 nmol, n = 8), into the Pn (vehicle, n = 8; 0.2 nmol, n = 8; 20 nmol, n = 14) and into mRt (vehicle, n = 4; 0.2 nmol, n = 8; 20 nmol, n =9). Arrows point to injection sites in the photographs. Numbers at the top of the drawings refer to anteroposterior stereotaxic coordinates from The Rat Brain atlas. Scale bar = 100 ␮m. mlf = medial longitudinal fasciculus; VTg = ventral tegmental nucleus; PnO = pontine nuclei, oral part; lfp = longitudinal fasciculus of the pons; Aq = aqueduct; xscp = decussation of the superior cerebellar peduncle.

points of injection were within the median raphe nucleus (MnR), whereas 30 were found at the pontine nuclei (Pn) and 21 at the mesencephalic reticular formation (mRt).

also remained unchanged after treatment with the other doses (Fig. 3). Non-ingestive and drinking behaviors were not affected by CLON treatment in the MnR. Correlations tests were performed and no association was found between variables.

3.1. Experiment 1—drug-dose effects on anxiety-like and ingestive and non-ingestive behaviors. Total risk assessment (TRA) frequency was affected by drugdose (F4,27 = 11.80; p = 0.00001). While the injection of the highest dose (20 nmol) of CLON into the MnR increased the frequency of total risk assessment, the lowest dose (0.2 nmol) decreased this behavior (Fig. 2). Anxiety-like parameters such as spatial-temporal variables and other ethological behaviors were not affected by CLON treatment in the MnR (Table 1). Food intake (F4,27 = 4.47; p = 0.006), feeding latency (F4,27 = 4.47; p = 0.006) and feeding frequency (F4,27 = 4.40; p = 0.007) were significantly affected by CLON in a dose related manner. While the treatment with the highest dose into the MnR induced hyperphagy, food intake remained unchanged after injection of lower doses (Fig. 3). The hyperphagic response evoked by the highest dose of CLON was accompanied by a reduction in the latency to start eating and an increase in feeding frequency but feeding duration was not affected (Fig. 3). The latency to start eating and feeding frequency

Fig. 2. Effects of saline (0) and clonidine into the median raphe nucleus on the frequency of total risk assessment, during 5 min of exposure to the elevated plusmaze. Values are mean ± S.E.M. of 6 animals/treatment, except for 20 nmol dose of CLON (n = 8). *p < 0.05 compared to saline, by one-way ANOVA and Duncan’s post hoc test.

74

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77

Table 1 Effects of saline and clonidine (CLON) into the median raphe nucleus on spatial-temporal and ethological variables, during 5 min of exposure to the elevated plus-maze. Variables

Saline (6)

CLON 0.2 (6)

% Open arm entries % Open arm time Enclosed entries Open entries Total entries Total head dipping Rearing Grooming

34.89 ± 9.33 23.05 ± 7,72 5.66 ± 1.42 4.16 ± 1.51 9.83 ± 2.57 8.16 ± 2.08 9.83 ± .79 2.5 ± .56

41.82 22.93 8.16 5.66 13.83 15.83 11.66 2.83

± ± ± ± ± ± ± ±

4.52 3.69 1.03 0.55 1.07 1.64 1.74 0.9

CLON 2 (6)

CLON 6 (6)

33.88 ± 7.89 12.02 ± 5.45 4.66 ± 1.42 3.33 ± 1.28 8. ± 2.67 8.66 ± 2.71 9.83 ± 2.15 2. ± 0.68

34.76 15.65 8.50 4.83 13.33 9.33 15.33 1.5

± ± ± ± ± ± ± ±

5.3 4.24 0.92 1.07 1.47 2.85 2.31 0.56

CLON 20 (8)

ANOVA

43.55 ± 7.29 20.02 ± 3.92 5.87 ± 1.27 4.50 ± 0.86 10.37 ± 1.68 11.625 ± 1.53 9.5 ± 1.73 3.87 ± 1.04

F4,27 = 0.42 F4,27 = 0.83 F4,27 = 1.70 F4,27 = 0.58 F4,27 = 1.46 F4,27 = 1.74 F4,27 = 1.74 F4,27 = 1.29

Values are mean ± S.E.M. Numbers in parentheses indicate the number of animals per group.

3.2. Experiment 2—CLON injections into the vicinity of the MnR

4. Discussion

The CLON doses (0.2 and 20 nmol) that had evoked changes in TRA frequency in the previous experiment were injected into areas adjacent to the MnR such as the Pn and the mRt. Whereas 0.2 nmol of CLON did not have any effect, TRA frequency was increased when 20 nmol were injected into the Pn or the mRt (F2,62 = 41.28; p = 10−6 ). This result was similar from the one obtained after the injection of the same dose into the MnR (Table 2). The dose of CLON that induced changes in food intake (20 nmol) was also injected into the Pn and mRt. This behavior was affected by the drug (F1,43 = 13.38; p = 0.006) and there was also significant interaction between the drug-dose and the injection site (F2,43 = 3.54; p = 0.037). Food intake increased only after CLON was injected into the MnR, but not when injected into the Pn or the mRt (Fig. 4).

The present study shows that while the spatial-temporal parameters remained unchanged after CLON injections into the MnR, the injection of the highest dose of this agonist increased the TRA frequency. As the risk assessment is positively correlated with anxiety-like behavior and thought to be one of the ethological parameters more sensitive for evaluation of anxiety-like behavior in the EPM test [62,66] we conclude that the activation of MnR ␣2 -adrenergic receptors induces anxiogenesis. As the activation of ␣2 -adrenergic receptors in the MnR inhibits the release of 5-HT in this nucleus projection sites [1,36,39] we suggest that the anxiogenesis may be mediated by a decrease in 5-HT release in limbic prosencephalic regions, such as the hippocampus and amygdala. More important, the injection of 8-OH-DPAT (0.5 ␮g/0.1 ␮l) into the MnR caused a marked decrease in 5-HT output in the hippocam-

Fig. 3. Effects of saline (0) and clonidine into the median raphe nucleus of free-feeding rats on food intake, feeding latency, feeding frequency and feeding duration, during 30 min of exposure to the food cage. Values are mean ± S.E.M. of 6 animals/treatment, except for 20 nmol dose of CLON (n = 8). *p < 0.05 compared to the other groups, by one-way ANOVA and Duncan’s post hoc test.

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77

75

Table 2 Effects of saline and clonidine (CLON) into the median raphe nucleus on the frequencies of non-ingestive variables of free-feeding rats, during 30 min of exposure to the food cage. Variables

Saline (6)

Locomotion Non-locomotion Immobility Rearing Grooming

59.16 99.5 34.16 22.16 19.66

± ± ± ± ±

13.79 14.32 10.6 5.68 13.07

CLON 0.2 (6) 70.66 103.00 23.00 26.66 15.66

± ± ± ± ±

CLON2 (6)

CLON 6 (6)

55.83 ± 10.4 98.5 ± 13.77 21.16 ± 6.7 25.66 ± 5.76 14.83 ± 2.67

51.33 96.83 28.50 19.33 18.83

19.71 8.33 8.46 7.13 3.07

± ± ± ± ±

10.14 6.8 8.15 4.83 2.24

CLON 20 (8) 51.62 76 15.12 21.25 17.35

± ± ± ± ±

9.79 6.18 5.92 3.58 2.42

ANOVA F4,27 = 0.36 F4,27 = 1.32 F4,27 = 0.88 F4,27 = 0.31 F4,27 = 0.79

Values are mean ± S.E.M. Numbers in parentheses indicate the number of animals per group.

pus (−41%) [10], which is, in the dorsal region, mostly innervated by fibers from the MnR [72,73]. Furthermore, the injection of 8-OHDPAT into the dorsal hippocampus evoked an anxiogenic-like effect [21,51,52,53], as well as into the MnR [21]. Altogether, these data support the results of the present study suggesting that the activation of MnR ␣2 -adrenergic receptors decreases the release of 5-HT from this nucleus and consequently evokes anxiogenesis, possibly changing the release of 5-HT into the hippocampus. The highest dose also increased the TRA frequency when injected in other structures near the MnR, such as the Pn and mRt. These results suggest that the anxiogenic-like effects of CLON are not specific to the MnR. Serotonergic neurons are also present in the Pn and mRt, although in smaller numbers than in the MnR [32]. Moreover, the presence of ␣2 -adrenergic receptors was demonstrated not only in the Pn but also in most sites of the brainstem [32]. However, to our knowledge, there are no data indicating the type of neurons that express adrenergic receptors in the Pn and mRt, these are unlikely to be 5-HT neurons. Furthermore, the functional role of the adrenergic receptors in these structures is still unknown. Whether similar adrenergic mechanisms controlling serotonin release found in the MnR exist in the Pn or the mRt deserves further investigation. In relation to the influence that the Pn and mRt neurons may have on behaviors, data showed that the Pn [45,59] and the mRt [54] play a role in anxiety-like behavior in rats. Additionally, the Pn also belongs to the central feeding circuitry [13,71], while feeding induced Fos expression in the mRt [33]. However, the literature does not show a relation between this nuclei and adrenergic drugs that act on 5-HT release and also the function of serotonergic neurotransmission upon these areas. This study also shows that whereas the highest dose of CLON injected into the MnR increased the TRA frequency, treatment with the lowest dose had the opposite effect. This fact might suggest that a reduction in 5-HT release mediated by the activation of MnR ␣2 -adrenergic receptors can also evoke anxiolytic-like behavior. In agreement with this, it has been reported that a reduced activity of MnR serotonergic neurons evokes anxiolytic-like behaviors

Fig. 4. Effects of clonidine (20 nmol) into the median raphe nucleus (MnR, n = 8), pontine nuclei (Pn/n = 14) and mesencephalic reticular formation (mRt/n = 9) of freefeeding rats on food intake, during 30 min of exposure to the food cage. Values are mean ± S.E.M. *p < 0.05 compared to the other groups, by two-way ANOVA and Duncan’s post hoc test.

in a number of anxiety-like models [3–6,18], including the EPM test [3,18] used in the present work. This paradoxical observation has been reported by others. Despite the fact that intra-MnR injections of 5-HT1A agonists evoke anxiolysis, infusions of the 5-HT1A receptor antagonist WAY-100635 into the MnR also produced an anxiolytic-like effect [12]. Additionally, lesions (5,7-DHT) of MnR5-HT neurons can cause anxiogenic [51], anxiolytic (3) or fail to alter anxiety-like indexes in the EPM [70]. Altogether, these data reveal that the MnR-5-HT mechanisms mediating anxiety-like behavior are conflicting and may vary according to the different drug-dose. This anxiety-like effect seems to be site-specific since the same dose of CLON failed to alter the TRA frequency when injected into the Pn or the mRt, adjacent to the MnR. Besides the anxiety-like effects, the data presented here also revealed that treatment with the highest dose of CLON into the MnR of free-feeding rats induced hyperphagia accompanied by a decrease in feeding latency and increase in feeding frequency. This hyperphagic effect may be attributed to the inhibitory influence of the ␣2 -adrenergic agonist upon MnR 5-HT neurons and reinforces our previous suggestion that the hyperphagic response evoked by AD injection into the MnR might be due to a reduced release of 5-HT in prosencephalic sites involved in food intake behavior [22]. Also, inhibition of MnR 5-HT neurons by local injection of 5-HT1A receptor agonist (8-OH-DPAT) increased feeding in satiated rats [28–30]. This evidence supports the suggestion that the ingestive effects caused after CLON injection may be mediated by alterations in the 5-HT release to MnR projection sites, such as hypothalamic nuclei. The changes in feeding behavior induced by CLON seem to be circumscribed to the MnR since no feeding responses were observed following injections into the Pn or the mRt. The feeding frequency and feeding latency patterns of this study were also similar to that evoked by AD injection into the MnR [22]. Changes in feeding frequency suggest alterations in a mechanism that initiates or terminates feeding [61]. Our data showed an increase in feeding frequency and a decrease in feeding latency, suggesting that MnR ␣2 -adrenergic receptors, in particular, control feeding initiation by means of 5-HT release. The satiation process involves the activation of NTS neurons such as the A2/C2 catecholaminergic cell groups [9]. The number of c-Fos positive cells increases in the NTS after treatment with anorexigenic peptides such as leptine [25] and cholecystokinin [55,80]. Furthermore, gastric distension increases Fos expression in tyrosine hydroxylase positive cells in the NTS [55,75,80]. As the processes of C2 neurons are generally directed sagitally, medially, and laterally along the ventricular floor and ventrally or medially toward the raphe [63], it is possible that this circuit is involved in the adrenergic control of feeding-related activity of MnR neurons. Interestingly, it has been demonstrated that sectioning of the vagus nerve, that courses from the gastro-intestinal tract to the nucleus of the solitary tract (NTS) in the caudal brainstem, abolishes the hyperphagic effect of ghrelin [14,46] a peptide that decreases the latency to feed and increases the number of meals without affecting meal size [26]. We suggest that adrenergic receptors in the MnR might contribute to the ghrelin effects on feeding behavior, since this peptide contributes to preprandial hunger and participates in meal initiation.

76

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77

It is estimated that 60% of MnR neurons are non-serotonergic [44]. It is possible that gabaergic or glutamatergic neurotransmission in the MnR [67] could be additional mechanisms involved in the hyperphagia evoked by CLON since intra-MnR injection of muscimol [42,57,76] and baclofen [76] or glutamatergic agonist (NMDA) [77] and antagonists (kynurenic acid and 2-amino-5phosphonovaleric acid) increased food intake in free-feeding rats [78]. Raphe/neurokinin pathways could also mediate the feeding effects induced by CLON treatment in the MnR since the activation of neurokinin-3 (NK-3) receptors located on MnR serotonin cell bodies, decreased food intake [57]. It is pertinent to note, however, the importance of serotonergic release on MnR projections sites, such as hippocampus or hypothalamus, after MnR ␣2 -adrenergic receptors activation on the control of anxiety-like and feeding behaviors. It is relevant to perceive that this study assessed the ingestive behavior after running the anxiety-like behavior test, which could affect the drug effect on feeding behavior. In a similar experiment, muscimol and baclofen were injected into the medial accumbens shell of free-feeding rats and the results showed hyperphagy and decrease in feeding latency; baclofen but not muscimol increased food length [47]. These responses were in accordance with other studies [7,40,60,68] that evaluated the feeding behaviors immediately after drug injection without assessing anxiety-like variables. In other experimental design that evaluated the feeding behaviors without setting the animals in the anxiety-like behavior test, AD was injected into the MnR of free-feeding rats and the effects were an increase in food intake, feeding frequency and decrease in latency to start feeding [22], results similar to the data presented here. Taken together, these data demonstrated that whether or not the animals are set in the EPM during 5 min before the feeding behavior evaluation, it does not seem to affect the feeding response evoked by drug. The changes induced by injecting CLON into the MnR on risk assessment behavior, as an anxiety-like variable, and the ones associated to food intake, can be understood by the same mechanism. The highest dose increased the TRA frequency, showing an anxiogenic-like effect, which could be caused by activation of ␣2 -adrenergic receptors in the MnR. The ␣2 -adrenergic receptors activation is known to reduce 5-HT release into terminal areas [1,2], such as the MnR serotonergic projections to the hippocampus [10]. In turn, the reduced availability of 5-HT to the hippocampus may evoke an anxiogenic-like effect. The clinical treatment is in line with this, as an increase in 5-HT availability has been shown to induct anxiolytic effects after the administration of selective 5-HT reuptake inhibitor [8,34]. The MnR ␣2 -adrenergic receptors activation also increased food intake, decreased the latency to start feeding and increased the feeding frequency. The serotonergic projections from the MnR to hypothalamus has been related to feeding response [16,17], that is, the decrease of 5-HT available in the hypothalamus after MnR CLON injection may reduce the inhibitory effects of 5-HT upon feeding behavior, evoking hyperphagy. Otherwise, the lowest dose decreased the TRA frequency, which indicates an anxiolytic-like effect. It is difficult to interpret the anxiolysis induced by the lowest dose of CLON into the MnR and we do not have a plausible explanation for it, however, the literature also shows an anxiolytic-like behavior after reduced activity of MnR serotonergic neurons of rats exposed to elevated plus-maze [3,18]. In addition, the fact that intra-MnR injections of 5-HT1A agonist and antagonist caused the same anxiolytic-like effect [12], represents a paradoxical observation of the 5-HT circuitry into MnR that controls anxiety. In summary, our data showed that the anxiolytic and anxiogenic-like effects mediated by activation of ␣2 -adrenergic receptors in the MnR seem to be dose related. The feeding response evoked by CLON injection in the MnR was similar to that induced by

injection of AD in the MnR, suggesting that the hyperphagia after AD treatment depends on ␣2 -receptors activation. The anxiolytic-like effect and feeding stimulation seem to be circumscribed to MnR. Furthermore, there are no functional associations between the MnR neural circuitries that control anxiety-like and feeding behaviors. Acknowledgments This study was financed by CNPq research grants to José Marino Neto (proc. 471888/03-6) and to Marta Aparecida Paschoalini (proc. 474762/01-7). References [1] Adell A, Artigas F. Regulation of the release of 5-hydroxytryptamine in the median raphe nucleus of the rat by catecholaminergic afferents. Eur J Neurosci 1999;11:2305–11. [2] Adell A, Celada P, Abellán MT, Artigas F. Origin and functional role of the extracellular serotonin in the midbrain raphe nuclei. Brain Res Rev 2002;39:154–80. [3] Andrade TGCS, Graeff FG. Effect of electrolytic and neurotoxic lesion of the median raphe nucleus on anxiety and stress. Pharmacol Biochem Behav 2001;70:1–14. [4] Andrade TGCS, Andrade CE, Avanzi V, Graeff FG. Effect of estradiol benzoate microinjection into the median raphe nucleus on contextual conditioning. Behav Brain Res 2009;24:112–6. [5] Andrews N, Hogg S, Gonzalez LE, File SE. 5-HT1A receptors in the median raphe nucleus of dorsal hippocampus may mediate anxiolytic and anxiogenic behaviours respectively. Eur J Pharmacol 1994;264:259–64. [6] Avanzi V, Brandão ML. Activation of somatodendritic 5-HT1A autoreceptors in the raphe nucleus disrupts the contextual conditioning in rats. Behav Brain Res 2001;126:175–84. [7] Basso AM, Kelley AE. Feeding induced by GABAA receptor stimulation within the nucleus accumbens shell: regional mapping and characterization of macronutrient and taste preference. Behav Neurosci 1999;113:324–36. [8] Baldwin DS, Tiwari N. The pharmacologic treatment of patients with generalized anxiety disorder: where are we now and where are we going? CNS Spectr 2009;14:5–12. [9] Berthoud HR. The vagus nerve, food intake and obesity. Regul Pept 2008;149:15–25. [10] Bonvento G, Scatton B, Claaustre Y, Rouquier L. Effect of local injection of 8-OHDPAT into the dorsal or median raphe nuclei on extracellular levels of serotonin in serotonergic projection areas in the rat brain. Neurosci Lett 1992;132:101–4. [11] Bortolozzi A, Artigas F. Control of 5-hydroxytryptamine release in the dorsal raphe nucleus by the noradrenergic system in rat brain: role of ␣adrenoreceptors. Neuropsychopharmacology 2003;28:421–34. [12] Canto-de-Souza A, Nunes-de-Souza RL, Rodgers RJ. Anxiolytic-like effect of way-100635 microinfusions into the median (but not dorsal) raphe nucleus in mice exposed to the plus-maze: Influence or prior test experience. Brain Res 2002;928:50–9. [13] Chen Z, Travers SP, Travers JB. Muscimol infusions in the brain stem reticular formation reversibly block ingestion in the awake rat. Am J Physiol Regul Integr Comp Physiol 2001;280:1085–94. [14] Crespo MA, González Matías LC, Gil Lozano M, Paz SF, Pérez MR, Gago EV, et al. Las hormonas gastrointestinales en el control de la ingesta de alimentos Gastrointestinal hormones in food intake control. Endocrinol Nutrición 2009;56:317–30. [15] Cryan JF, Page ME, Lucki I. Noradrenergic lesions differentially alter the antidepressant-like effects of reboxetine in a modified forced swim test. Eur J Pharmacol 2002;436:197–205. [16] Currie PJ, Fletcher PJ, Coscina DV. Administration of 8-OH-DPAT into the midbrain raphe nuclei: effects on medial hypothalamic NE-induced feeding. Am J Physiol 1994;266:1645–51. [17] Currie PJ, Coscina DV. Diurnal variations in the feeding response to 8-OHDPAT injected into the dorsal or median raphe. Neuroreport 1993;4:1105–7. [18] De Almeida RM, Giovenardi M, Charchat H, Lucion AB. 8-OH-DPAT in the median raphe nucleus decreases while in the medial septal area it may increase anxiety in female rats. Neurosci Biobehav Rev 1998;23:259–64. [19] De Vry J, Schreiber R. Effects of selected serotonin 5-HT(1) and 5-HT(2) receptor agonists on feeding behavior: possible mechanisms of action. Neurosci Biobehav Rev 2000;24:341–53. [20] Dominguez R, Cruz-Moralesa SE, Carvalhoa MC, Xaviera M, Brandão ML. Sex differences in serotonergic activity in dorsal and median raphe nucleus. Physiol Behav 2003;80:203–10. [21] Dos Santos L, Andrade TGCS, Zangrossi Jr H. 5-HT1A receptors in the dorsal hippocampus mediate the anxiogenic effect induced by the stimulation of 5-HT neurons in the median raphe nucleus. Eur Neuropsychopharmacol 2008;18:286–94. [22] Dos Santos RLD, Mansur SS, Steffens SM, Faria MS, Marino Neto J, Paschoalini MA. Food intake increased after injection of adrenaline into the median raphe nucleus of free-feeding rats. Behav Brain Res 2009;197:411–6. [23] Dourish CT, Hutson PH, Curzon G. Characteristics of feeding induced by the serotonin agonist 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT). Brain Res Bull 1985;15:377–84.

S.S. Mansur et al. / Behavioural Brain Research 212 (2010) 71–77 [24] Ebenezer IS, Arkle MJ, Tite RM. 8-Hydroxy-2-(di-n-propylamino)-tetralin inhibits food intake in fasted rats by an action at 5-HT1A receptors. Methods Find Exp Clin Pharmacol 2007;29:269–72. [25] Emond EE, Ladenheim, Schwartz LGJ, Moran T. Leptin amplifies the feeding inhibition and neural activation arising from a gastric nutrient preload. Physiol Behav 2001;72:123–8. [26] Faulconbridge LF, Cummings DE, Kaplan JM, Grill HJ. Hyperphagic effects of brainstem ghrelin administration. Diabetes 2003;52:2260–5. [27] File SE, Hyde JRG, Mc Leod NK. 5,7-Dyhydroxytryptamine lesion of dorsal and median raphe nuclei and performance in the social interaction test of anxiety and in a home cage aggression test. J Affect Disord 1979;1:115–22. [28] Fletcher PJ. Dopamine receptor blockade in nucleus accumbens or caudate nucleus differentially affects feeding induced by 8-OH-DPAT injected into dorsal or median raphe. Brain Res 1991;552:181–9. [29] Fletcher PJ, Coscina DV. Injecting 5-HT into the PVN does not prevent feeding induced by injecting 8-OH-DPAT into the raphe. Pharmacol Biochemi Behav 1993;46:487–91. [30] Funk D, Li Z, Fletcher J, Le AD. Effects of injections of 8-hydroxy-2-(dinpropylamino) tetralin or muscimol in the median raphe nucleus on c-fos mRNA in the rat brain. Neuroscience 2005;131:475–9. [31] Hall H, Lundkvist C, Halldin C, Farde L, Pike VW, McCarron JA, et al. Autoradiographic localization of 5HT1A receptors in the post-mortem human brain using [3H] WAY-100635 and [11C] WAY 100635. Brain Res 1997;745:96–108. [32] Happe HK, Coulter CL, Gerety ME, Sanders JD, O’Rouke M, Bylund DB, et al. Alpha-2 adrenergic receptor development in rat CNS: Na autoradiographic study. Neuroscience 2004;123:167–78. [33] Hironaka S, Shirakawa T, Toki S, Kinoshita K, Oguchi H. Feeding-induced c-fos expression in the nucleus of the solitary tract and dorsal medullary reticular formation in neonatal rats. Neurosci Lett 2000;293:175–8. [34] Hood Sd, Hince DA, Davies SJ, Ar gyropoulos S, Robinson H, Potokar J, et al. Effects of acute tryptophan depletion in serotonin reuptake inhibitorremitted patients with generalized anxiety disorder. Psychopharmacology 2010;2:223–32. [35] Holmes A. Genetic variation in cortico-amygdala serotonin function and risk for stress-related disease. Neurosci Biobehav Rev 2008;32:1293–314. [36] Hopwood SE, Stamford JA. Noradrenergic modulation of serotonin release in rat dorsal and median raphe nuclei via ␣1 and ␣2 adrenoceptors. Neuropharmacology 2001;41:433–42. [37] Hornung JP. The human raphe nuclei and the serotonergic system. J Chem Neuroanat 2003;26:331–43. [38] Judge SJ, Young RL, Gartside S. GABA-A receptor modulation of 5-HT neuronal firing in the median raphe nucleus: implications for the action of anxiolytics. Eur Neuropsychopharmacol 2006;16:612–9. [39] Judge SJ, Gartside SE. Firing of 5-HT neurons in the dorsal and median raphe nucleus in vitro shows differential alpha1-adrenoceptor and 5-HT1A receptor modulation. Neurochem Int 2006;48:100–7. [40] Kelley AE, Baldo BA, Pratt WE, Will MJ. Corticostriatal-hypothalamic circuitry and food motivation: integration of energy, action and reward. Physiol Behav 2005;86:773–95. [41] Kia HK, Miquel MC, Brisorgueil MJ, Daval G, Riad M, El Mestikawy S, et al. Immunocytochemical localization of serotonin 1A receptors in the rat central nervous system. J Comp Neurol 1996;365:289–305. [42] Klitenick MA, Wirtshafter D. Elicitation of feeding, drinking, and gnawing following microinjections of muscimol into the median raphe nucleus of rats. Behav Neural Biol 1989;51:436–41. [43] Le AD, Harding S, Funk D, Juzytsch W, Shaham Y. The role of alpha-2 adrenoceptors in stress-induced reinstatement of alcohol seeking and alcohol self administration in rats. Psychopharmacology 2005;179:366–73. [44] Leger L, Wiklund L. Distribution and numbers of indoleamine cell bodies in the cat brainstem determined with Falck-Hillarp fluorescence histochemistry. Brain Res Bull 1982;9:245–51. [45] Leri F, Franklin KB. Learning impairments caused by lesions to the pedunculopontine tegmental nucleus: an artifact of anxiety? Brain Res 1998;807:187–92. [46] Le Roux CW, Neary NM, Halsey TJ, Small CJ, Martinez-Isla AM, Ghatei MA. Ghrelin does not stimulate food intake in patients with surgical procedures involving vagotomy. J Clin Endocrinol Metab 2005;90:4521–4. [47] Lopes APF, Cunha IC, Steffens SM, Ferraz A, Vargas JC, Lima TCM, et al. GABAA and GABAB agonist microinjections into medial accumbens shell increase feeding and induce anxiolysis in an animal model of anxiety. Behav Brain Res 2007;184:142–9. [48] López-Alonso VE, Mancilla-Díaz JM, Rito-Domingo M, González-Hernández B, Escartín-Pérez RE. The effects of 5-HT1A and 5-HT2C receptor agonists on behavioral satiety sequence in rats. Neurosci Lett 2007;416:285–8. [49] Maidel S, Lucinda AM, Aquino VW, Faria MS, Paschoalini MA. The adrenaline microinjection into the median raphe nucleus induced hypophagic effect in rats submitted to food restriction regimen. Neurosci Lett 2007;422:123–7. [50] Mokler DJ, Dugal JR, Hoffman JM, Morgane PJ. Functional interrelations between nucleus raphé dorsalis and nucleus raphé medianus: a dual probe microdialysis study of glutamate-stimulated serotonin release. Brain Res Bull 2009;78:132–8.

77

[51] Netto SM, Silveira R, Coimbra NC, Joca SRL, Guimarães FS. Anxiogenic effect of median raphe nucleus lesion in stressed rats. Progr Neuro-Psychopharmacol Bio Psychiatry 2002;26:1135–41. [52] Nunes-de-Souza RL, Canto-de-Souza A, Rodgers RJ. Anxiolytic-like effect of WAY-100635 microinfusions into ventral but not dorsal hippocampus in mice. J Psychopharmacol 1999;A57(Suppl. A). [53] Nunes-de-Souza RL, Canto-de-Souza A, da-Costa M, Fornari RV, Graeff FG, Pela IR. Anxiety-induced antinociception in mice: effects of systemic and intraamygdala administration of 8-OH-DPAT and midazolam. Psychopharmacology 2000;150:300–10. [54] Olds ME, Olds J. Effects of anxiety-relieving drugs on unit discharges in hippocampus, reticular midbrain, and pre-optic area in the freely moving rat. Int J Neuropharmacol 1969;8:87–103. [55] Olson BR, Freilino M, Hoffman GE, Stricker EM, Sved A, Verbalis JG. c-Fos expression in rat brain and brainstem nuclei in response to treatments that alter food intake and gastric motility. Mol Cell Neurosci 1993;4:93–106. [56] Ottoni EB. Etholog 2.2: a tool for the transcription and timing of behavior observation sessions. Behav Res Methods Instrum Comput 2000;32:446–9. [57] Paris JM, Mitsushio H, Lorens SA. Intra-midbrain raphe injections of the neurokinin-3 agonist senktide inhibit food and water intake in the rat. Pharmacol Bioch Behav 1991;38:223–6. [58] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. sixth ed. NewYork: Academic Press and Elsevier; 2007. [59] Podhorna J, Franklin KB. Lesions of the pedunclopontine tegmental nucleus increase anxiety in rats. Neuroreport 1998;9:1783–6. [60] Reynolds SM, Berridge KC. Fear and feeding in the nucleus accumbens shell: rotrocaudal segregation of GABA-elicited defensive behavior versus eating behavior. J Neurosci 2001;21:3261–70. [61] Ritter RC, Epstein AN. Control of meal size by central noradrenergic action. Proc Natl Acad Sci USA 1975;72:3740–3. [62] Rodgers RJ, Cao BJ, Dalvi A, Holmes A. Animal models of anxiety: an ethological perspective. Braz J Med Bio Res 1997;30:289–304. [63] Ruggiero DA, Ross CA, Anwar M, Park DH, Joh TH, Reis DJ. Distribution of neurons containing phenylethanolamine N-methyltransferase in medulla and hypothalamus of rat. J Comp Neurol 1985;239:127–54. [64] Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nature 2000;404:661–71. [65] Shim I, Javaid J, Wirtshafter D. Dissociation of hippo- campal serotonin release and locomotor activity following pharmacological manipulations of the median raphe nucleus. Behav Brain Res 1997;89:191–8. [66] Setem J, Pinheiro AP, Motta VA, Morato S, Cruz APM. Ethopharmacological analysis of 5-HT ligands on the rat elevated plus-maze. Pharmacol Biochem Behav 1999;62:515–21. [67] Sotelo C, Cholley B, Mestikawy SE, Gozlan H, Hamon M. Direct immunohistochemical evidence of the existence of 5-HT1A autoreceptors on serotonergic neurons in the midbrain raphe nuclei. Eur J Neurosci 1990;2: 1144–54. [68] Stratford TR, Kelly AE. GABA in the nucleus accumbens shell participates in the central regulation of feeding behavior. J Neurosci 1997;17:4434–40. [69] Takase LF, Nogueira MI. Patterns of Fos activation in rat raphe nuclei during feeding behavior. Brain Res 2008;1200:10–8. [70] Thomas H, Fink H, Sohr R, Voits M. Lesion of the median raphe nucleus: a combined behavioural and microdialysis study in rats. Pharmacol Biochem Behav 2000;65:15–21. [71] Trojniar W, Staszewska M. Bilateral lesions of the pedunculopontine tegmental nucleus affect feeding induced by electrical stimulation of the ventral tegmental area. Acta Neurobiol Exp 1995;55:201–6. [72] Vertes RP, Crane AM. Distribution, quantification and morphological characteristics of serotonin-immunoreactive cells of the supralemniscal nucleus (B9) and pontomesencephalic reticular formation in the rat. J Comp Neurol 1997;378:411–24. [73] Vertes RP, Fortin WJ, Crane AM. Projections of the median raphe nucleus in the rat. J Comp Neurol 1999;407:555–82. [74] Walther DJ, Bader M. A unique central tryptophan hydroxylase isoform. Biochem Pharmacol 2003;66:1673–80. [75] Willing AE, Berthoud HR. Gastric distension-induced c-fos expression in catecholaminergic neurons of rat dorsal vagal complex. Am J Physiol 1997;272:59–67. [76] Wirtshafter D. The control of ingestive behavior by the median raphe nucleus. Appetite 2001;36:99–10561. [77] Wirtshafter D, Krebs JC. Control of food intake by kainate/quisqualate receptors in the median raphe nucleus. Psychopharmacology 1990;101:137–41. [78] Wirtshafter D, Trifunovic R. Stimulation of ingestive behaviors following injections of excitatory amino acid antagonists into the median raphe nucleus. Pharmacol Biochem Behav 1988;30:529–33. [79] Woods SC, Schwartz MW, Baskin DG, Seeley RJ. Food intake and the regulation of body weight. Annu Rev Psychol 2000;51:255–77. [80] Zittel TT, Glatzle J, Kreis ME, Starlinger M, Eichner M, Raybould HE, et al. c-Fos protein expression in the nucleus of the solitary tract correlates with cholecystokinin dose injected and food intake in rats. Brain Res 1999;846:1–11.