Brain Research Bulletin 69 (2006) 356–364

Research report

Serotoninergic activation of the basolateral amygdala andmodulation of tonic immobility in guinea pig

Christie Ramos Andrade Leite-Panissi a,∗, Aline Aparecida Ferrarese a,Ana Luisa Bernardes Terzian a, Leda Menescal-de-Oliveira b

a Department of Morphology, Stomatology and Physiology, Dental School of Ribeirao Preto, University of Sao Paulo, 14040-904 Ribeirao Preto, SP, Brazilb Department of Physiology, Medicine School of Ribeirao Preto, University of Sao Paulo, 14049-900 Ribeirao Preto, SP, Brazil

Received 6 December 2005; received in revised form 3 February 2006; accepted 13 February 2006Available online 7 March 2006

Abstract

Tonic immobility (TI), also known as death feigning or animal hypnosis, is a reversible state of motor inhibition that is not only triggered bypostural inversion and/or movement restraining maneuvers but also by repetitive stimulation and pressure on body parts. Evidence has demonstratedthat the basolateral nucleus of the amygdala (BLA) is particularly associated with defensive behavior that involves the emotional states of fearaaraoBrTd©

K

1

rifptocbvc

0d

nd anxiety. In addition, some reports have demonstrated that serotonin (5-HT) released in the amygdala is increased during states of stress andnxiety, principally in the BLA. In the present study, we investigated the effects of serotonergic activation of the BLA on the duration of TI. Theesults showed that the microinjection of 5-HT (3.0 �g) into the BLA decreased the duration of TI. Similarly, the administration of a 5-HT1A

gonist (0.1 �g of 8-hydroxy-dipropylaminotretalin) or 5-HT2 agonist (0.1 �g of �-methyl-5-HT) into the BLA reduced the TI duration. The effectf 5-HT2 agonist was reversed by pretreatment with a dose that had no effect per se (0.01 �g) of ketanserin (5-HT2 receptor antagonists) into theLA. Moreover, the activation of 5-HT1A and 5-HT2 receptors in the BLA did not alter the spontaneous motor activity in the open field test. The

esults of the present study indicate that the serotonergic system of the BLA possibly produces a reduction in fear and/or anxiety that reduces theI duration in guinea pigs, but this is not due to increased spontaneous motor activity induced by serotonergic activation, which might affect TIuration non-specifically.

2006 Elsevier Inc. All rights reserved.

eywords: Amygdala; Serotonin; Defensive behavior; Fear; Anxiety; Open field

. Introduction

Tonic immobility (TI) behavior is well established as a fearesponse, which can be elicited in a wide range of vertebrates andnvertebrates [31], and is established most easily through someorm of restraint or physical inversion [17,51]. It can be a com-onent of antipredator behavior and is the last resource used byhe prey to reduce the probability of continued attack on the partf the predator, thus being of adaptive value [61]. This responsean last from a few seconds to several hours and is characterizedy the temporary suppression of the righting response, reducedocalization, wavy flexibility, hypertonicity of the skeletal mus-les, Parkinsonian-like muscle tremors in the extremities, altered

∗ Corresponding author. Tel.: +55 16 3602 4124; fax: +55 16 3633 0999.E-mail address: christie@forp.usp.br (C.R.A. Leite-Panissi).

electroencephalographic patterns, and changes in heart rate, res-piration and core temperature [28,31].

A recent study was realized to validate the TI responsein the guinea pig as a pharmacological model for the detec-tion of anxiolytic compounds [46]. In the cited report, theauthors demonstrated that the serotonergic, noradrenergic andneurokinin systems are involved in mediating or modulating TIbehavior. However, more results are required to determine thepredictive validity of the TI model for the detection of anx-iolytic and/or antidepressant drug activity. A correspondencebetween human defensive behavior and fear and anxiety-relateddefense patterns of non-human mammals is possible to establish[3]. It has been shown that rodent defensive behaviors elicitedby threat, such as flight, freezing, defensive threat, defensiveattack and risk assessment, are responsive to drugs that areeffective against anxiety or depressive disorders in humans [4].In addition, selective serotonin reuptake inhibitors are first-line

361-9230/$ – see front matter © 2006 Elsevier Inc. All rights reserved.oi:10.1016/j.brainresbull.2006.02.007

C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364 357

treatment for most anxiety disorders, but traditional animal mod-els of anxiety have failed to show the anxiolytic effect of theserotonergic system [6].

With respect to the role of the serotonin (5-HT) mechanismin TI, systemic studies have demonstrated that low (0.24 mg/kg,i.v.) doses of serotonin (5-HT) produced significant increasesin the duration of TI, while high (21.8 mg/kg, i.v.) doses onlyproduced a slight decrease in duration in chickens [24]. Indeed,intraventricular administration of 5-HT in chickens producedan increase in TI duration [21], while a similar procedure withrabbits produced a decrease in the duration of TI [23]. Thesedifferent results have been attributed to differences in the routeof drug administration or the species used. Though, with sys-temic injections the 5-HT levels in all central nervous systemwere altered. Another recent study showed that the action of5-HT inside the ventrolateral and dorsal periaqueductal graymatter (PAG) seems to be biphasic and dose-dependent becausein these regions the microinjection of low doses (0.1 �g) of 5-HT increased TI, while high doses (1, 3 and 6 �g) decreased thisbehavior. The PAG 5-HT1A and 5-HT2 receptors have differentroles in the modulation of TI, since the 5-HT1A and 5-HT2 ago-nists, respectively, increased and decreased the duration of TI[44].

Several studies have been conducted to determine the cen-tral nervous system regions that may control the TI response.Klemm [31] proposed that TI may be considered to be a reflexwomrttlbtntgid

ld[ef(dahsotceit

the ventrolateral PAG [34]. In fact, the amygdaloid complexreceives highly processed sensory information from all modali-ties through its lateral and basolateral nuclei. In turn, these nucleiproject to the CEA, which then projects to a variety of hypotha-lamus nuclei, including the PAG and brainstem target areas thatdirectly mediate specific signs of fear and anxiety [9].

Some reports have demonstrated that 5-HT release in theamygdala is increased during states of stress and anxiety, prin-cipally in the BLA [30,53]. Additionally, the amygdala receivesa prominent innervation from serotonergic terminals originat-ing from neurons of the dorsal raphe nucleus [37], and neuronalinhibition by different manipulations of the serotonergic neuronsfrom the dorsal raphe nucleus have produced anxiolytic effectsin several animal models [20]. Although there is a high den-sity of 5-HT receptors within the amygdaloid complex, the roleof these receptors in modulating fear and anxiety presents con-tradictory effects. Consequently, the activation of post-synaptic5-HT1A receptors in the BLA may produce anxiogenic effectsin the social interaction test [19]. In contrast, the same acti-vation with 8-hydroxy-dipropylaminotretalin (8-OH-DPAT; 5-HT1A agonist) of BLA receptors promoted an anxiolytic effectin the elevated T-maze [64].

Based on the findings described above, the present studywas designed to investigate the effects of serotonergic activa-tion of the BLA on the duration of TI. To this end, we inves-tigated whether microinjection of serotonin, 8-OH-DPAT (5-Hnid

2

2

owttaaUfa

2

abihtgcwpzwe

hose expression appears to be related to the activation of a poolf interneurons in the reticular formation of the brain stem thatay influence descending neurons which inhibit the motor neu-

ons of the spinal medulla. Although, experiments with brainransection [31] have shown that the brain stem is fundamen-al for the expression of this behavior, other regions such asimbic system and the neocortex may also modulate this immo-ility response. In this way, previous reports have demonstratedhat the parabrachial nucleus [42], PAG [43,44], hypothalamusuclei [45] and amygdaloid complex [34–36] are involved inhe modulation of TI in guinea pigs. Again, LeDoux [33] sug-ested that the amygdala provides a more refined control of thentensity and timing of the defense reaction integrated in theiencephalic–mesencephalic and periventricular core.

The amygdaloid complex is a small, almond-shaped structureocated deep within the temporal lobe and lies at the center ofefense system involved in conditioned and unconditioned fear9,10]. Lesions of the amygdala block the acquisition and/or thexpression of many aspects of conditioned and unconditionedear responses [10,12,18]. Within the amygdala, the basolateralBLA) and central (CEA) nuclei are particularly associated withefensive behaviors that involve emotional states of fear andnxiety [9,57]. Preliminary studies carried out in our laboratoryave shown that the cholinergic, opioidergic and GABAergicystems of the CEA have an inhibitory action on the durationf TI. In addition, the GABAergic circuit of the CEA has aonic inhibitory influence on the duration of TI, since bicu-ulline administrated into the CEA promoted an increase in TIpisode [35]. Furthermore, we have shown that the CEA partic-pates in the modulation of TI response and also in antinocicep-ion by means of a functional and anatomical connection with

T1A receptor agonist), �-methyl-5-HT (5-HT2 receptor ago-ists) and ketanserin (5-HT2 receptor antagonists) into the BLAn different groups of guinea pigs produced an alteration in TIuration.

. Methods

.1. Animals

Adult male guinea pigs (Cavia porcellus, University of Sao Paulo, Campusf Ribeirao Preto, Brazil) weighing 400–500 g (n = 83) were kept in Plexiglasall cages (56 cm × 17 cm × 39 cm, five guinea pigs per cage) in a room main-

ained at 24 ± 1 ◦C, on a 12-h light cycle, with free access to water and foodhroughout the experimental period. The experiments were carried out in compli-nce with the recommendations of SBNeC (Brazilian Society of Neurosciencend Behavior), with the approval (No. 05.1.586.53.2) of the Animal Care andse Committee of the University of Sao Paulo, Campus of Ribeirao Preto and

ollowing their ethical recommendations. All efforts were made to minimizenimal suffering.

.2. Tonic immobility recording

Each guinea pig was submitted to five control maneuvers of TI inductionnd the duration of the episodes was recorded. Induction of TI was attemptedy holding the guinea pig around the thorax with the hands, quickly invertingt, and pressing it down into a shaped plywood trough (25 cm long × 15 cmigh). The pressure applied by the hands of the experimenter was proportionalo the resistance offered by the guinea pig to the restraining maneuver. When theuinea pig stopped moving, the experimenter slowly withdrew his hands and ahronometer was activated to measure the duration (in seconds) of the response,hich ended when the guinea pig resumed the upright position. If the guineaig did not become motionless within 60 s, the episode was recorded as havingero duration. For group analysis, the mean of five episodes per guinea pigas considered. The minimum acceptable average duration of TI in the control

pisode was TI ≥ 45 s.

358 C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364

2.3. Surgical procedures

One day after the control TI episode, the guinea pigs were anesthetized by anintramuscular injection of ketamine (100 mg/kg) plus xylazine (14 mg/kg) andplaced in a stereotaxic apparatus (David-Kopf Instruments, USA) with the buc-cal piece 21.4 mm below the interauricular line, and one guide cannula (14 mmin length and 0.6 mm in outer diameter, prepared from a hypodermic needle) wasimplanted into the left hemisphere toward the basolateral nucleus of the amyg-dala (BLA). According to the Rossner [52] atlas for guinea pigs, the stereotaxiccoordinates for the placement of the guide cannula implanted toward the BLAwere 3.4 mm caudal to the bregma, 6.0 mm lateral to the midline and 9.0 mmbelow the cortical surface. The guide cannula was lowered to a depth of 1 mmabove the target regions and fixed to the skull by means of self-polymerizingresin and an additional anchoring screw. After recovery from anesthesia undera heat lamp, the animals were returned to the housing area.

2.4. Experimental groups of tonic immobility

After recovering from surgery for 5–6 days, the guinea pigs were divided intoseven experimental groups to evaluate the effects of the drugs. In groups 1 (n = 6),2 (n = 8) and 3 (n = 10), the guinea pigs were first submitted to a TI session afterthe administration of 0.2 �l of 0.1% ascorbic acid solution for vehicle control.The following day the same guinea pigs were microinjected with different dosesof 5-HT 0.1, 0.3 and 3.0 �g for groups 1, 2 and 3, respectively.

To evaluate the effect of 5-HT1A and 5-HT2 agonists receptors, after recov-ery from surgery, the guinea pigs were submitted to a TI session without drugadministration for post-operative control (SHAM) and then they received differ-ent doses (0.01 and 0.1 �g) of 8-OH-DPAT (group 4, n = 10) or �-methyl-5-HT(group 5, n = 10) on the subsequent 2 days. The dose-related effect of the antag-onist of 5-HT2, ketanserin, was studied in group 6 (n = 5). Thus, after recoveringfa(o7(f

2

�

trdwr0Hwiott

2

dwdws

cdj

over a period of 1 min, and the needle was left in place for an additional 40 s toavoid reflux.

2.7. Histological verification

After the end of the experiments, the animals were anesthetized with 2,2,2-tribroethanol (2.5%, 10 ml/kg, i.p.; Aldrich, WI, USA), and intracardially per-fused with saline followed by 10% formalin. The brains were removed andfixed in 10% formalin. The material was then submitted to routine histologicalprocessing and sections were observed under a microscope to determine thelocations of the stimulated sites according to the Rossner atlas [52]. Only theguinea pigs whose microinjections reached the target structure were used fordata analysis.

2.8. Statistical analysis

The results of TI behavior are reported as the means ± standard error ofthe mean (S.E.M.) of the mean duration of the five episodes of TI. Due tovariability and bias in the TI duration data, a natural logarithmic transformationwas performed on all duration scores prior to statistical analysis. The results oflocomotor activity are reported as the means ± S.E.M. of the number of squarescrossed during 5 min in the open field test. The data were analyzed by analysisof variance for repeated measures (ANOVA). The degree of freedom of therepeated measure (treatment) was corrected by the Huynh–Feldt ε parameter.The Tukey test was used to determine the difference between treatments, withthe level of significance set at P < 0.05.

3. Results

Fig. 1 illustrates the microinjection sites made in the basolat-e

bi3aswmkp

bTHFcHnPt

(tA(nw0tO

rom surgery, the guinea pigs were submitted to the SHAM episode and 24 hfter the same guinea pigs/rodents received two different doses of ketanserin0.01 and 0.03 �g) on 2 consecutive days. To evaluate whether an effective dosef ketanserin (0.01 �g) was able to block the effect of �-methyl-5-HT, in group(n = 6) 24 h after the SHAM episode the guinea pigs received �-methyl-5-HT

0.1 �g) and ketanserin (0.01 �g) 5 min prior to �-methyl-5-HT (0.1 �g) on theollowing day. All drugs were administrated in a volume of 0.2 �l.

.5. Motor activity

To exclude the possibility that serotonergic activation by 8-OH-DPAT and-methyl-5-HT administration reduced TI non-specifically by increasing spon-

aneous activity, motor activity was measured after the microinjection of rep-esentative doses of serotonergic agonists that significantly decreased the TIuration. Thus, in three experimental groups the guinea pigs were implantedith unilateral cannulae toward the basolateral nucleus of the amygdala. After

ecovering from surgery for 5–6 days, the guinea pigs were microinjected with.9% saline (n = 8), 8-OH-DPAT (n = 10, at a dose of 0.1 �g) or �-methyl-5-T at a dose of 0.1 �g (n = 11) and the open field test was realized. The testas performed in a wood box (60 cm × 60 cm × 60 cm) with the ground divided

nto 16 equal squares of 15 cm each. The guinea pig was placed in the centerf the box and the locomotor activity was evaluated by direct observation ofhe number of squares crossed per minute during 5 min, immediately after thereatment.

.6. Drugs and microinjection procedure

The drugs used, 5-hydroxytryptamine (5-HT, serotonin), 8-hydroxy-ipropylaminotretalin (8-OH-DPAT), �-methyl-5-HT and ketanserin (KET),ere obtained from Sigma (St. Louis, MO, USA). The solutions were freshlyissolved in 0.9% saline containing 0.1% ascorbic acid and sodium bicarbonateas added to adjust the pH to 7.4. The doses employed were based on previous

tudy data [44].The microinjections were performed with a Hamilton microsyringe (10 �l)

onnected to a PE-10 polyethylene catheter, which in turn was coupled to a thinental needle (0.3 mm o.d.; 1 mm longer than the guide cannula). The microin-ections were made on consecutive days. A volume of 0.2 �l was microinjected

ral nucleus of the amygdala for all experimental groups.The results showed that the serotonergic activation of the

asolateral amygdala decreased the TI duration in guinea pign all studied groups. In fact, the administration of 5-HT at.0 �g (Fig. 2C), 8-OH-DPAT at a dose of 0.1 �g (Fig. 3A)nd �-metyhl-5-HT at a dose of 0.1 �g (Fig. 3B) produced aignificant reduction in TI behavior. In contrast, the treatmentith ketanserin (5-HT2 receptor antagonist; Fig. 4A) did not pro-ote any alteration in TI duration, however pretreatment with

etanserin into the BLA blocked the decrease in TI durationromoted by �-metyhl-5-HT (Fig. 4B).

Repeated measures ANOVA reveled a significant differenceetween treatments (F2,18 = 6.65, P < 0.001) in group 3. Theukey post hoc test showed a significant difference between 5-T 3.0 �g, and the control and 0.1% ascorbic acid (AA; P < 0.05;ig. 2C). However, no difference was observed between theontrol and AA. In contrast, in groups 1 and 2, the effect of 5-T microinjection at doses of 0.1 and 0.3 �g into the BLA didot differ significantly from the control or the AA (F2,14 = 8.36,> 0.05; F2,10 = 0.96, P > 0.05, ANOVA; groups 1 and 2, respec-

ively; Fig. 2A and B).The microinjection of 8-OH-DPAT at a dose of 0.1 �g

group 4) into the BLA also significantly decreased the dura-ion of TI episodes (Fig. 3A). In group 4, repeated measuresNOVA showed a significant difference between treatments

F3,24 = 26.13, P < 0.001). The Tukey test demonstrated a sig-ificant difference between 8-OH-DPAT 0.1 �g when comparedith the control (P < 0.05), SHAM (P < 0.05) and 8-OH-DPAT.01 �g (P < 0.05). However, no difference was found withinhe groups control, SHAM and microinjection of 0.01 �g 8-H-DPAT into the BLA.

C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364 359

Fig. 1. Schematic drawing of frontal sections obtained at representative levelsof the guinea pig amygdala. (A) The filled circles (�) represent the sites whereserotonin (5-HT) 0.1 �g was microinjected; the open circles (©) represent thesites where 5-HT 0.3 �g was microinjected and the filled lozenges (�) representthe sites where 5-HT 3.0 �g was microinjected. (B) The open star (�) depictthe sites where 8-OH-DPAT at doses of 0.01 and 0.1 �g was microinjected; thefilled squares (�) indicate the sites of microinjection of �-methyl-5-HT (a-methyl-5-HT) at doses of 0.01 and 0.1 �g was microinjected; the open squares(�) represent the sites where ketanserin (KET) at doses of 0.01 and 0.03 �gwas microinjected; the filled triangles (�) indicate the microinjection sites of �-methyl-5-HT (a-methyl-5-HT) at a dose of 0.1 �g and �-methyl-5-HT precededby KET were microinjected. (C) The open triangles (�) indicate the sites wheresaline was microinjected; the filled triangles (�) depict the microinjection sitesof 8-OH-DPAT; the filled stars (�) depict the microinjection sites of �-methyl-5-HT (a-methyl-5-HT). The number of points in the figure is less than the totalnumber of guinea pigs (n = 83) due to several overlaps. All microinjections were

The effect of microinjecting �-methyl-5-HT into the BLAis shown in Fig. 3B. In group 5, the microinjection of �-methyl-5-HT at a dose of 0.1 �g into the BLA caused a signifi-cant reduction in TI episodes. These findings were supportedby repeated measures ANOVA, which revealed a significantdifference between treatments (F3,24 = 30.46, P < 0.001). TheTukey test showed a significant difference between �-methyl-5-HT 0.1 �g when compared with the control (P < 0.05), SHAM(P < 0.05) and �-methyl-5-HT 0.01 �g (P < 0.05). However, thecontrol, SHAM and �-methyl-5-HT 0.01 �g treatments did notdiffer among themselves.

In group 6, the microinjection of ketanserin at doses of 0.01and 0.03 �g into the BLA did not alter TI duration (F3,12 = 2.96,P > 0.05, ANOVA; Fig. 4A). However, the previous adminis-tration of ketanserin 0.01 �g into the BLA was able to blockthe decrease in TI duration (F3,18 = 10.03, P < 0.001, ANOVA;Fig. 4B) induced by treatment with �-methyl-5-HT at a doseof 0.1 �g (group 7). The Tukey post hoc test revealed a sig-nificant difference between �-methyl-5-HT 0.1 �g when com-pared with the control, SHAM (P < 0.05) and the treatment withketanserin followed by �-methyl-5-HT 0.1 �g (P < 0.05). How-ever, no difference was observed among the treatments precededby ketanserin, control and SHAM.

Finally, the unilateral microinjection of 8-OH-DPAT or �-methyl-5-HT at a dose of 0.1 �g into the BLA did not alterspontaneous motor activity in the open field test (F = 0.88,Pti1w�

4

gapmatstaotpsst

mAp(

2,26> 0.05, ANOVA; Fig. 5) when compared with the animals

hat received an administration of saline into the BLA. Thus,n the control group the mean number of squares crossed was.9 ± 0.6, in the group that was administered 8-OH-DPAT thisas 3.6 ± 1.0 and in the group that received a microinjection of-methyl-5-HT at a dose of 0.1 �g this was 3.4 ± 1.0.

. Discussion

The results of the present study indicate that the serotoner-ic system of the BLA probably produces a reduction in fearnd/or anxiety, consequently reducing the TI duration in guineaigs, but apparently this is not due to increased spontaneousotor activity induced by serotonergic activation, which may

ffect TI duration non-specifically. Many studies have shownhe importance of emotional factors, fear in particular, in theusceptibility to, or duration of, TI episodes [17,31]. Consis-ent with this view, we may hypothesize that a reduction in fearnd anxiety by the activation of the 5-HT1A or 5-HT2 receptorsf the BLA [11,27,64] could be one of the factors involved inhe reduction of TI duration in guinea pigs, as observed in theresent study. In addition, several clinical placebo-controlledtudies have consistently shown that drugs assumed to facilitateerotonin neurotransmission prevent anxiety and are effective inreating anxiety disorders [15].

ade on the left side, but some groups are illustrated on the right for clarity.bbreviations: BLA, basolateral amygdala; CEA, central amygdala; LPA, lateralosterior amygdala; OT, optic tract. (A and B) The TI experimental groups andC) the groups of the open field test.

360 C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364

Fig. 2. Effect of different doses of 5-HT on the duration of tonic immobilityepisodes. Mean TI durations before (CONT) and after acid ascorbic (AA, 0.1%)and after serotonin (5-HT) at different doses. (A) Group 1 (n = 6) where theguinea pigs received 5-HT at a dose of 0.1 �g/0.2 �l into the basolateral nucleusof the amygdala (BLA). (B) Group 2 (n = 8) where the guinea pigs were microin-jected with 5-HT at a dose of 0.3 �g/0.2 �l into the BLA. (C) Group 3 (n = 10)where the guinea pigs received 5-HT at a dose of 3.0 �g/0.2 �l into the BLA.*P < 0.05 compared with the control and after microinjection of AA by the Tukeytest. The vertical bars indicate the S.E.M.

Supporting the involvement of the serotonergic system in themodulation of TI behavior, Olsen et al. [46] demonstrated that anincrease in 5-HT release induced by fenfluramine s.c. caused adecrease in TI duration in the guinea pig. These authors suggestthat this effect may be due to the stimulation of post-synaptic5-HT1A receptors, as fenfluramine significantly increased 5-HTin the synapse, and 5-HT1A receptor agonism induced by bus-pirone or 8-OH-DPAT also attenuated TI duration. Similarly,a recent study has demonstrated that the microinjection of 5-HT into the ventrolateral and dorsal PAG seems to be biphasic

Fig. 3. Effect of 5-HT1A or 5-HT2 agonist receptors microinjected into thebasolateral nucleus of the amygdala (BLA) of guinea pigs on the duration oftonic immobility episodes. (A) Mean TI durations before (CONT), after surgeryin order to implant the guide cannula (SHAM) and after 8-OH-DPAT at dosesof 0.01 �g/0.2 �l and 0.1 �g/0.2 �l microinjected on 2 consecutive days into theBLA (n = 9). (B) Mean TI durations before (CONT), after surgery in order toimplant the guide cannula (SHAM) and after �-methyl-5-HT (a-methyl-5-HT)at doses of 0.01 �g/0.2 �l and 0.1 �g/0.2 �l microinjected on 2 consecutive daysinto the BLA (n = 9). *P < 0.05 compared to the control and SHAM by the Tukeytest. The vertical bars indicate the S.E.M.

and dose-dependent. The microinjection of low doses (0.1 �g)of 5-HT into the PAG increased the duration of TI, while highdoses (1, 3 and 6 �g) decreased this behavior [44]. Thus, thesame neurotransmitter can modulate the same behavior in oppo-site ways depending on the neural structure. Davis et al. [13]showed that 5-HT infused into the lateral ventricle in rats pro-duced a dose-dependent depression of the acoustic startle reflex.When infused into the spinal cord, serotonin produced a dose-dependent increase in startle.

Several in vivo microdialysis studies examined the effect ofconditioned fear on extracellular serotonin concentrations insome brain areas, including the amygdala, principally in theBLA [16,29,30]. Thus, exposure to context paired with footshocks increased extracellular serotonin concentration in theamygdala [29]. Simultaneous observations of extracellular sero-tonin concentration and freezing indicated that serotonin levelsdo not increase while the rats freeze and that the serotonin levelincreased after the rats recovered from freezing [22]. Anotherrecent report has shown that increased serotonin concentrationin the amygdala decreased conditioned fear [27]. Taken together,

C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364 361

Fig. 4. Duration of tonic immobility episodes. (A) Mean TI durations before(CONT), after surgery in order to implant the guide cannula (SHAM) and aftermicroinjection of ketanserin (KET) at doses of 0.01 �g/0.2 �l and 0.03 �g/0.2 �lon 2 consecutive days into the basolateral nucleus of the amygdala (n = 5).(B) Mean TI durations before (CONT), after surgery in order to implant theguide cannula (SHAM) and after �-methyl-5-HT (a-methyl-5-HT) at the dose0.1 �g/0.2 �l preceded by KET at doses of 0.01 �g/0.2 �l into the BLA on 2consecutive days (n = 7). *P < 0.05 compared with the control, SHAM and afterKET followed by �-methyl-5-HT (a-methyl-5-HT) by the Tukey test. The ver-tical bars represent S.E.M.

these results support the present findings, making it is possibleto suggest that serotonergic activation of the amygdala occursafter the expression of conditioned or unconditioned fear anddoes not induce fear.

Fig. 5. Effect of 5-HT1A or 5-HT2 agonist receptors microinjected into thebasolateral nucleus of the amygdala (BLA) of guinea pigs on motor activity(counts) for 5 min in the open field test. Motor activity was counted 1 min afterthe microinjection of 0.9% saline (SAL, 0.2 �l, n = 8), 8-OH-DPAT (5-HT1A

agonist at doses of 0.1 �g/0.2 �l, n = 10) or �-methyl-5-HT (a-methyl-5-HT;5-HT2 agonist at doses of 0.1 �g/0.2 �l, n = 10).

Although there is a relatively high density of 5-HT receptorswithin the amygdaloid complex [49], the role of theses recep-tors in modulating fear and anxiety causes some controversy. Infact, the amygdaloid complex is innervated by 5-HT-containingfibers originating from the dorsal raphe nucleus [37]. Serotoninis widely distributed within the amygdala, including profuseterminal labeling in the central, basolateral and cortical groups.Moreover, the amygdala has a differential distribution of 5-HTreceptor subtypes. Hence 5-HT1A receptors are found predomi-nantly in the central nucleus, whereas 5-HT2, 5-HT3 and 5-HT6receptors are found predominantly in the basolateral complex[49]. Consequently, infusions of the 5-HT1A receptor agonist,8-OH-DPAT into the lateral and basolateral nuclei of the amyg-dala decreased punished response, indicating anxiogenic actions[26]. The same 5-HT1A receptor agonist in the BLA [19] causedan overall reduction in the levels of social investigation, whichalso shows the anxiogenic effect of 5-HT1A receptor activationin the BLA. It is interesting to note that in the Gonzalez et al. [19]study the administration within BLA was bilateral, with the useof a volume of 0.5 �l, which can reach the central nucleus, as wellas the lateral nucleus, in addition to the BLA. Indeed, Wise andHoffman [63] suggested that intracerebral drug injection witha volume of 0.5 �l can diffuse one or more millimeters fromthe infusion site. Furthermore, BLA has only a low density of5-HT1A receptors, the highest levels being found in the centralnucleus [47,48]. In contrast, in our experiments, the microin-jptotao

tiijPmsepnAspttait

ti[p

ection of 8-OH-DPAT was unilateral with a volume of 0.2 �l,romoting a serotonergic activation of a smaller area limited tohe BLA, the subject of the present study. In agreement withur results, Zangrossi et al. [64] showed that limited administra-ion of 8-OH-DPAT into the BLA promoted impaired inhibitoryvoidance of the open arms of the T-maze, indicating an anxi-lytic effect.

With respect to 5-HT2 receptors, peripheral activation ofhese receptors has been shown to either decrease [25,46], orncrease [25,62] TI duration. In agreement with our results, butnvestigating the serotonergic circuitry in the PAG, the microin-ection of 5-HT2 agonist into the ventrolateral and dorsolateralAG decreased the duration of TI, an effect reversed by pretreat-ent with an ineffective dose of ketanserin [44]. In the present

tudy, ketanserin, an unspecific 5-HT2 antagonist, blocked theffect of �-methyl-5-HT on TI duration when microinjectedrior to �-methyl-5-HT into the BLA. This showed the specificature of the response and the participation of 5-HT2 receptors.lthough the BLA is one of the amygdaloid nuclei richest in

erotonergic innervation [49], the microinjection of ketanseriner se into the BLA did not change TI duration, suggesting thathe serotonergic system did not participate in a tonic manner inhe organization of this defensive behavior. Similarly, Stein etl. [59] demonstrated that the administration of ketanserin alonento the BLA had no effect on the discharge rate of neurons ofhe same site.

Anatomical studies of the amygdaloid complex have shownhat the BLA is a major input nucleus of the amygdala, receivingnformation from all sensory modalities and from the forebrain1]. The BLA consists of two major classes of neurons, largeyramidal cells and smaller non-pyramidal cells [39]. In addi-

362 C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364

tion, projection neurons of the BLA are immunopositive forglutamate, suggesting that they utilize this excitatory amino acidas their primary neurotransmitter [41] and stellate cells in theBLA are GABAergic interneurons [40].

Considerable evidence has demonstrated that 5-HT depressesexcitatory synaptic transmission and depolarization-evokedCa2+ influx in the basolateral amygdala via 5-HT1A receptors[7]. Indeed, a 5-HT1A agonist significantly inhibited the firingrate of the neurons within the basolateral amygdala [59]. How-ever, the low levels of 5-HT1A receptor expression observed inthe BLA suggest that the 5-HT-mediated inhibition of BLA cellfiring may not result from a direct activation of post-synaptic 5-HT1A receptors but, rather, by an indirect inhibition of glutamaterelease via activation of presynaptic 5-HT1A receptors [5]. More-over, expression of 5-HT2 receptors on interneurons of the BLAsuggest that inhibition of cell firing also may be facilitated bya direct excitation of GABAergic interneurons [59,60]. Further-more, recent studies in the amygdala suggest that acute serotoninrelease could directly activate GABAergic interneurons via theactivation of 5-HT2 receptors and increase the frequency ofinhibitory synaptic events in projection neurons, also in thisregion [60]. It is possible that in our results the microinjection of8-OH-DPAT or �-methyl-5-HT was able to increase the inhibi-tion of glutamate release in BLA projection neurons promotinganxiolytic effect evaluated by the decrease in TI response. Inagreement with our hypothesis, bilateral injection of citalo-pbdtmstneotot

fcthmijaogaeisaus

Another plausible explanation of the reduced duration of TIepisodes after serotonergic stimulation of the BLA could be viathe activation of the BLA–PAG connection. Da Costa Gomezet al. [8] demonstrated that PAG cells respond to BLA electri-cal stimulation and that the majority of these cells are locatedin the dorsolateral and lateral columns of the PAG. Further-more, the stimulation of the dorsal PAG produces defensivereactions associated with fear and anxiety, such as hyperten-sion, tachycardia and freezing [2,14]. In this case, the sero-tonergic activation of the BLA could reduce the firing rateof the projection neurons in this region and, subsequently,reduce glutamatergic release in the dorsal and lateral PAG,which could reduce the expression of defensive behavior mod-ulated by the dorsal and lateral PAG. Reinforcing the partic-ipation of the glutamate receptors in the dorsal and lateralPAG in the control of defensive behavior, Bittencourt et al.[2] demonstrated that microinjection of NMDA into the dor-sal and lateral columns of the PAG evoked exophthalmia andimmobility, the responses which characterize freezing behav-ior.

In this study, we showed that the microinjection of 5-HT, or ofthe agonist to 5-HT1A and 5-HT2 receptors promoted a decreasein TI duration, possibly due to a reduction in fear and anxi-ety, but not due to increased spontaneous motor activity, whichmay affect TI behavior non-specifically. Our findings supportthe view that increased activation of the serotonergic system oftmcteettn

A

t(t

R

ram (selective serotonin reuptake inhibitors) into the amygdalaefore testing reduced conditioned freezing significantly [27]emonstrated the anxiolytic effect of the serotonergic system ofhe amygdala. Again, stimulation of the GABAergic system with

uscimol, a GABAA receptor agonist, or by the facilitation ofynaptic GABAergic transmission by means of the administra-ion of benzodiazepines into the lateral posterior or basolateraluclei of the amygdala [54,55,58], promoted a potent anxiolyticffect in rats. This effect was blocked by previous administrationf bicuculline methiodide into the lateral or basolateral nuclei ofhe amygdala [56], supporting the hypothesis of the involvementf the amygdaloid complex GABAergic system in the modula-ion of emotional responses.

Several findings support the hypothesis that the BLA per-orms part of its functions by means of a connection with theentral nucleus of the amygdala [8,50]. Furthermore, the cen-ral nucleus of the amygdala projects to the PAG [34], to theypothalamus [32] and the brainstem areas [32] that directlyediate specific signs of fear and anxiety. In the present study,

t is possible that the inhibition of glutamate release in BLA pro-ection neurons induced by serotonergic activation could haveltered the excitatory tonus on neurons of the CEA. In fact,ur previous report [35] showed that the cholinergic, opioider-ic and GABAergic system of the CEA presented an inhibitoryction on the duration of TI response. Additionally, Macedot al. [38] showed that lesions of the serotonergic terminalsn the basolateral complex decreased the regulatory role oferotonergic mechanisms normally exerted at this level of themygdala, and proposed that disruption of the inhibitory reg-lation existent in the BLA might greatly amplify fear andtress.

he BLA is the mechanism of anxiolytic action. Perhaps, the TIodel should be used in conjunction with other measurements to

larify the effects of anxiolytic drugs, such as serotonergic drugs,o provide new insights into our understanding of the pathogen-sis and pathophysiology of fear and anxiety disorders. Furtherxperiments are required to understand the interplay betweenhe different 5-HT receptor subtypes and other neurotransmit-er systems, like the GABAergic circuit, within the basolateralucleus of the amygdala.

cknowledgements

We would like to thank Mrs. Nadir Martins Fernandes for herechnical assistance. This work was supported by the FAPESPGrant nos. 03/01471-5, 04/09642-6 to A.A.F. and 04/13060-2o A.L.B.T.) and the CNPq.

eferences

[1] D.G. Amaral, J.L. Price, A. Pitkanen, S.T. Carmichael, Anatomical orga-nization of the primate amygdaloid complex, in: J.P. Aggleton (Ed.), TheAmygdala: Neurobiological Aspects of Emotions, Memory and MentalDysfunction, Willy-Liss, 1992, pp. 1–66.

[2] A.S. Bittencourt, A.P. Carobrez, L.P. Zamprogno, S. Tufik, L.C. Schen-berg, Organization of single components of defensive behaviors withindistinct columns of periaqueductal gray matter of the rat: role of N-methyl-d-aspartatic acid glutamate receptors, Neuroscience 125 (2004)71–89.

[3] D.C. Blanchard, G. Griebel, R.J. Blanchard, Mouse defensive behaviors:pharmacological and behavioral assays for anxiety and panic, Neurosci.Biobehav. Rev. 25 (2001) 205–218.

[4] R.J. Blanchard, G. Griebel, B. Guardiola-Lemaitre, M.M. Brush, J. Lee,D.C. Blanchard, An ethopharmacological analysis of selective activa-

C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364 363

tion of 5-HT1A receptors: the mouse 5-HT1A syndrome, Pharmacol.Biochem. Behav. 57 (1997) 897–908.

[5] D.H. Bobker, J.T. Williams, Serotonin agonist inhibit synaptic poten-tials in the rat locus ceruleus in vitro via 5-hydroxytrytamine1A and5-hydroxytrytamine1B receptors, J. Pharmacol. Exp. Ther. 250 (1989)37–43.

[6] F. Borsini, J. Podhorna, D. Marazziti, Do animal models of anxiety pre-dict anxiolytic-like effects of antidepressants? Psychopharmacol. (Berl.)163 (2002) 121–141.

[7] L.L. Cheng, S.J. Wang, P.W. Gean, Serotonin depresses excitatory synap-tic transmission and depolarization-evoked Ca2+ influx in rat basolateralamygdala via 5-HT1A receptors, Eur. J. Neurosci. 10 (1998) 2163–2172.

[8] T.M. Da Costa Gomez, S.D. Chandler, M.M. Behbehani, The role ofthe basolateral nucleus of the amygdala in the pathway between theamygdala and the periaqueductal gray in the rat, Neurosci. Lett. 214(1996) 5–8.

[9] M. Davis, The role of the amygdala in fear and anxiety, Annu. Rev.Neurosci. 15 (1992) 353–375.

[10] M. Davis, The role of the amygdala in emotional learning, Int. Rev.Neurobiol. 36 (1994) 225–266.

[11] M. Davis, R.L. Commissaris, J.V. Cassella, S. Yang, L. Dember, T.P.Harty, Differential effects of dopamine agonists on acoustically and elec-trically elicited startle responses: comparison to effects of strychnine,Brain Res. 371 (1986) 58–69.

[12] M. Davis, W.A. Falls, S. Campeau, M. Kim, Fear-potentiated startle:a neural and pharmacological analysis, Behav. Brain Res. 58 (1993)175–198.

[13] M. Davis, D.I. Strachan, E. Kass, Excitatory and inhibitory effects ofserotonin on sensorimotor reactivity measured with acoustic startle, Sci-ence 209 (1980) 521–523.

[

[

[

[

[

[

[

[

[

[

[

[

[26] H. Hodges, S. Green, B. Glenn, Evidence that the amygdala is involvedin benzodiazepine and serotonergic effects on punished respondingbut not on discrimination, Psychopharmacol. (Berl.) 92 (1987) 491–504.

[27] T. Inoue, X.B. Li, T. Abekawa, Y. Kitaichi, T. Izumi, S. Nakagawa,T. Koyama, Selective serotonin reuptake inhibitor reduces conditionedfear through its effect in the amygdala, Eur. J. Pharmacol. 497 (2004)311–316.

[28] E.G. Jones, Neurotransmitters in the cerebral cortex, J. Neurosurg. 65(1986) 135–153.

[29] H. Kawahara, Y. Kawahara, M. Yoshida, H. Yokoo, M. Nishi, M. Tanaka,Serotonin release in the rat amygdala during exposure to emotionalstress, in: A. Tanaka, G. Curson (Eds.), Serotonin in the Central NervousSystem and Periphery, Elsevier, Amsterdam, 1995, pp. 89–96.

[30] H. Kawahara, M. Yoshida, H. Yokoo, M. Nishi, M. Tanaka, Psycholog-ical stress increases serotonin release in the rat amygdala and prefrontalcortex assessed by in vivo microdialysis, Neurosci. Lett. 162 (1993)81–84.

[31] W.R. Klemm, Neurophysiologic studies of the immobility reflex (“Ani-mal hypnosis”), Neurosci. Res. 4 (1971) 165–212.

[32] J.E. Krettek, J.L. Price, Amigdaloid projections to subcortical structureswithin the basal forebrain and brainstem en the rat and cat, J. Comp.Neurol. 178 (1978) 225–254.

[33] J.E. LeDoux, Emotion, memory and the brain, Sci. Am. 270 (1994)50–57.

[34] C.R.A. Leite-Panissi, N.C. Coimbra, L. Menescal-de-Oliveira, Thecholinergic stimulation of the central amygdala modifying the tonicimmobility response and antinociception in guinea pigs depends onthe ventrolateral periaqueductal gray, Brain Res. Bull. 60 (2003) 167–178.

[35] C.R.A. Leite-Panissi, L. Menescal-de-Oliveira, Central nucleus of the

[

[

[

[

[

[

[

[

[

[

[

14] A. Depaulis, K.A. Keay, R. Bandler, Longitudinal neuronal organizationof defensive reactions in the midbrain periaqueductal gray region of therat, Exp. Brain Res. 90 (1992) 307–318.

15] E. Erikkson, M. Humble, Serotonin in psychiatric pathophysiology. Areview of data from experimental and clinical research, in: R. Pohl, S.Gershon (Eds.), Progress in Basic and Clinical Pharmacology, vol. 3,Karger, Basel, 1990, pp. 66–119.

16] C. Fernandes, N. Andrews, S.E. File, Diazepam withdrawal increases[3H]-5-HT release from rat amygdaloid slices, Pharmacol. Biochem.Behav. 49 (1994) 359–362.

17] G.G. Gallup Jr., Tonic immobility: the role of fear and predation, Psy-chol. Rec. 27 (1977) 41–61.

18] L.E. Goldstein, A.M. Rasmusson, B.S. Bunney, R.H. Roth, Role ofthe amygdala in the coordination of behavioral, neuroendocrine, andprefrontal cortical monoamine responses to psychological stress in therat, J. Neurosci. 16 (1996) 4787–4798.

19] L.E. Gonzalez, N. Andrews, S.E. File, 5-HT1A and benzodiazepinereceptors in the basolateral amygdala modulate anxiety in the socialinteraction test, but not in the elevated plus-maze, Brain Res. 732 (1996)145–153.

20] G. Griebel, 5-Hydroxytryptamine-interacting drugs in animal models ofanxiety disorders: more than 30 years of research, Pharmacol. Ther. 65(1995) 319–395.

21] C.T. Harston, D.H. Sibley, G.G. Gallup Jr., W.B. Wallman, Effectsof intraventricular injections of imipramine and 5-hydroxytryptramineon tonic immobility in chickens, Bull. Psychon. Soc. 8 (1976) 403–405.

22] S. Hashimoto, T. Inoue, T. Koyama, Effects of conditioned fear stresson serotonin neurotransmission and freezing behavior in rats, Eur. J.Pharmacol. 378 (1999) 23–30.

23] D.C. Hatton, R. Tankle, T. Lanthorn, M. Meyer, Serotonin and tonicimmobility in the rabbit, Behav. Biol. 24 (1978) 97–100.

24] C.W. Hennig, Biphasic effects of serotonin on tonic immobility indomestic fowl, Pharmacol. Biochem. Behav. 12 (1980) 519–523.

25] C.W. Hennig, W.C. Steinhoff, S.A. Braughler, Tonic immobility and theserotoninergic system in chickens: differential blockade of receptors bycyproheptadine and cinanserin, Physiol. Behav. 44 (1988) 15–20.

amygdala and the control of tonic immobility in guinea pigs, Brain Res.Bull. 58 (2002) 13–19.

36] C.R.A. Leite-Panissi, C.R. Monassi, L. Menescal-de-Oliveira, Role ofthe amygdaloid nuclei in the modulation of tonic immobility in guineapigs, Physiol. Behav. 67 (1999) 717–724.

37] Q.P. Ma, G.F. Yin, M.K. Ai, J.S. Han, Serotonergic projections from thenucleus raphe dorsalis to the amygdala in the rat, Neurosci. Lett. 134(1991) 21–24.

38] C.E. Macedo, V.M. Castilho, M.A. de Souza e Silva, M.L. Brandao,Dual 5-HT mechanisms in basolateral and central nuclei of amyg-dala in the regulation of the defensive behavior induced by electricalstimulation of the inferior colliculus, Brain Res. Bull. 59 (2002) 189–195.

39] A.J. McDonald, Cell types and intrinsic connection of the amygdala,in: J.P. Aggleton (Ed.), The Amygdala: Neurobiological Aspects ofEmotion, Memory, and Mental Dysfunction, Wiley-Liss, 1992, pp. 67–96.

40] A.J. McDonald, J.R. Augustine, Localization of GABA-like immunore-activity in the monkey amygdala, Neuroscience 52 (1993) 281–294.

41] A.J. McDonald, J.C. Pearson, Coexistence of GABA and peptideimmunoreactivity in non-pyramidal neurons of the basolateral amygdala,Neurosci. Lett. 100 (1989) 53–58.

42] L. Menescal-de-Oliveira, A. Hoffmann, The parabrachial region as apossible region modulating simultaneously pain and tonic immobility,Behav. Brain Res. 56 (1993) 127–132.

43] C.R. Monassi, C.R.A. Leite-Panissi, L. Menescal-de-Oliveira, Ventro-lateral periaqueductal gray matter and the control of tonic immobility,Bull. Brain Res. 50 (1999) 201–208.

44] C.R. Monassi, L. Menescal-de-Oliveira, Serotonin 5-HT2 and 5-HT1Areceptors in the periaqueductal gray matter differentially modulate tonicimmobility in guinea pig, Brain Res. 1009 (2004) 169–180.

45] L. Oliveira, A. Hoffmann, L. Menescal-de-Oliveira, The lateral hypotha-lamus in the modulation of tonic immobility in guinea pigs, Neuroreport8 (1997) 3489–3493.

46] C.K. Olsen, S. Hogg, M.D. Lapiz, Tonic immobility in guinea pigs:a behavioural response for detecting an anxiolytic-like effect? Behav.Pharmacol. 13 (2002) 261–269.

364 C.R.A. Leite-Panissi et al. / Brain Research Bulletin 69 (2006) 356–364

[47] A. Pazos, J.M. Palacios, Quantitative autoradiographic mapping of sero-tonin receptors in the rat brain. I. Serotonin-1 receptors, Brain Res. 346(1985) 205–230.

[48] F. Radja, A.M. Laporte, G. Daval, D. Verge, H. Gozlan, M. Hamon,Autoradiography of serotonin receptor in the central nervous system,Neurochem. Int. 18 (1991) 1–15.

[49] D.G. Rainnie, Serotonergic modulation of neurotransmission in the ratbasolateral amygdala, J. Neurophysiol. 82 (1999) 69–85.

[50] D.G. Rainnie, Inhibitory and excitatory circuitries in amygdala nuclei:a synopsis of session II, Ann. N. Y. Acad. Sci. 985 (2003) 59–66.

[51] S.C. Ratner, Comparative aspects of hypnosis, in: J.E. Gordon (Ed.),Handbook of Clinical and Experimental Hypnosis, Macmilian, NewYork, 1967, pp. 550–587.

[52] W. Rossner, Stereotaktischer Hirnatlas vom Meerchwlinchen, PallaVelag, Munich, 1965.

[53] L.E. Rueter, B.L. Jacobs, A microdialysis examination of serotoninrelease in the rat forebrain induced by behavioral/environmental manip-ulations, Brain Res. 739 (1996) 57–69.

[54] S.K. Sanders, A. Shekhar, Blockade of GABAA receptors in the regionof the anterior basolateral amygdala of rats elicits increases in heart rateand blood pressure, Brain Res. 567 (1991) 101–110.

[55] S.K. Sanders, A. Shekhar, Regulation of anxiety by GABAA receptorsin the rat amygdala, Pharmacol. Biochem. Behav. 52 (1995) 701–706.

[56] J. Scheel-Kruger, E.N. Petersen, Anticonflit effect of the benzodiazepinesmediated by a GABAergic mechanism in the amygdala, Eur. J. Pharma-col. 82 (1982) 115–116.

[57] A. Shekhar, T.J. Sajdyk, D.R. Gehlert, D.G. Rainnie, The amygdala,panic disorder, and cardiovascular responses, Ann. N. Y. Acad. Sci. 985(2003) 308–325.

[58] K. Shibata, Y. Kataoka, Y. Gomita, S. Ueki, Localization of the site ofthe anticonflict action of benzodiazepines in the amygdaloid nucleus ofrats, Brain Res. 234 (1982) 442–446.

[59] C. Stein, H. Davidowa, D. Albrecht, 5-HT(1A) receptor-mediated inhi-bition and 5-HT(2) as well as 5-HT(3) receptor-mediated excitation indifferent subdivisions of the rat amygdala, Synapse 38 (2000) 328–337.

[60] G.E. Stutzmann, J.E. LeDoux, GABAergic antagonists block theinhibitory effects of serotonin in the lateral amygdala: a mechanism formodulation of sensory inputs related to fear conditioning, J. Neurosci.19 (1999) RC8.

[61] R.W. Thompson, R.K.R. Foltin, R.J. Boylan, A. Sweet, C.A. Graves,C.E. Lovwitf, Tonic immobility in japanese quauil can reduced theprobability of sustained attack by cats, Anim. Learn. Behav. 9 (1981)145–149.

[62] L.B. Wallnau, G.D. Bordash, P. Corso Jr., The effects of tryptophan andmanipulations of serotonergic receptors on tonic immobility in chickens,Pharmacol. Biochem. Behav. 14 (1981) 463–468.

[63] R.A. Wise, D.C. Hoffman, Localization of drugs reward mechanisms byintracranial injections, Synapse 10 (1992) 247–263.

[64] H. Zangrossi Jr., M.B. Viana, F.G. Graeff, Anxiolytic effectof intra-amygdala injection of midazolam and 8-hydroxy-2-(di-n-propylamino)tetralin in the elevated T-maze, Eur. J. Pharmacol. 369(1999) 267–270.