Brain Research 960 (2003) 201–208www.elsevier.com/ locate/brainres

Research report

C orticotropin-releasing factor in the dorsal raphe nucleus regulatesactivity of lateral septal neurons

a , a,c b c*Earl Thomas , Luise Pernar , Irwin Lucki , Rita J. ValentinoaDepartment of Psychology, Bryn Mawr College, 101 N. Merion Avenue, Bryn Mawr, PA 19010,USAbDepartment of Psychiatry, University of Pennsylvania, 415 Curie Blvd., Philadelphia, PA 19104,USA

cDepartment of Pediatrics, Children’ s Hospital of Philadelphia, Civic Center Blvd., Philadelphia, PA 19104,USA

Accepted 16 October 2002

Abstract

Corticotropin-releasing factor (CRF) has substantial effects on brain serotonergic activity, especially in limbic structures related tostress and anxiety. For example, relatively low doses of CRF administered into the dorsal raphe nucleus (DRN) decrease DRN unitactivity and serotonin release in the lateral septum (LS), a limbic target of the DRN. In contrast, higher doses of CRF tend to be excitatoryon both endpoints. The present experiment sought to establish the functional connection between CRF effects in the DRN and the ultimateeffect on activity in the LS as a terminal region. We recorded the effects of CRF (3, 10, 30 and 100 ng in 100 nl of artificial cerebrospinalfluid) administered into the DRN upon LS unit activity. In general, the lower doses of CRF (3 and 10 ng) had a facilitatory effect on LSunit activity, peaking at about 15–20 min post-injection. The higher doses had a more complex effect with an early suppression of unitresponding maximizing at about 5 min followed by a facilitatory rebound, especially at the 100 ng dose, maximizing at about 20 min.Taken with previous studies demonstrating an inhibitory effect of 5-HT on neuronal activity in LS, the findings suggest that CRFregulation of the DRN is translated to changes in LS activity. This effect may underlie certain coping behaviors in response to stress. 2002 Elsevier Science B.V. All rights reserved.

Theme: Neurotransmitters, modulators, transporters, and receptors

Topic: Peptides, anatomy and physiology

Keywords: Corticotropin-releasing factor; Dorsal raphe; 5-Hydroxytryptamine; Limbic system; Lateral septum; Serotonin; Single-unit; Stress

1 . Introduction [16,22,27–29] and that the brain serotonergic system mayplay a key role in the behavioral effects of CRF [11].

Corticotropin-releasing factor (CRF) functions both as a Anatomical studies indicate that the DRN, a major siteneurohormone, promoting the release of ACTH and as a of 5-HT cell bodies projecting to the forebrain, is targetedneurotransmitter, playing a significant role in the mediation by CRF. Thus, the DRN contains CRF binding sites [7],of the stress response in the brain [18,38]. Brain regions and CRF receptor mRNA is expressed in DRN [4,26].subserving the stress-related effects of CRF include limbic CRF immunoreactive terminals are localized in DRNstructures such as the central nucleus of the amygdala [19], [16,30,34]. Recent microdialysis studies have revealed thathypothalamic nuclei such as the paraventricular nucleus relatively low doses of CRF administered i.c.v. or directly[24], and brainstem structures such as locus coeruleus [37]. into the DRN inhibit 5-HT release in two forebrainRecent research has indicated that CRF has substantial terminal regions of DRN, the striatum and the lateraleffects on brain serotonergic activity, especially in limbic septum, whereas higher doses either increased or did notstructures mediating the response to stress and anxiety alter 5-HT release in these areas [27,29]. The effects of

CRF on DRN neurons mirror its effects on 5-HT release.Thus, low doses of CRF administered i.c.v., comparable to*Corresponding author. Tel.:11-610-526-5013; fax:11-610-526-those that diminish 5-HT release in the forebrain regions,7476.

E-mail address: ethomas@brynmawr.edu(E. Thomas). decreased single-cell discharge rate in DRN [16,29].

0006-8993/02/$ – see front matter 2002 Elsevier Science B.V. All rights reserved.PI I : S0006-8993( 02 )03882-9

202 E. Thomas et al. / Brain Research 960 (2003) 201–208

Moreover, CRF administered directly into DRN had a (Harlan, Frederick MD) weighing 250–300 g. Animalsbiphasic effect on DRN neuronal discharge, with lower were individually housed in a light and temperaturedoses (1.0, 3.0 and 10 ng) inhibiting DRN discharge and a controlled colony. The animals were maintained on a 12-hhigher dose (30 ng) facilitating neuronal discharge [16]. light /dark cycle. They were provided with ad lib food andTogether these studies suggest that the effects of CRF on water.forebrain 5-HT release are mediated by its effects on DRNneuronal activity. 2 .2. Electrophysiology and CRF administration

An extensive body of evidence has implicated brainCRF in the behavioral manifestations of stress and anxiety. Ovine CRF was generously provided by Dr Jean RivierCRF administered i.c.v. has been shown to be anxiogenic of the Clayton Foundation for Peptide Biology, The Salkin a number of animal models. Thus, CRF increases the Institute, La Jolla, CA. Animals were anesthetized with

21effects of punishment in the conflict test [3,6] and de- urethane (1.4 g kg i.p.) and placed in a stereotaxiccreases the time spent in the open arms of the elevated apparatus. Body temperature was maintained at 37–388Cplus-maze [3]. Intracranial CRF also potentiates the acous- with a heating pad. To approach the DRN for CRFtic startle response in a manner similar to the effects of a injection a hole was drilled on the midline 7.8 mmconditioned fear stimulus [20]. posterior to bregma. The dura and saggital sinus were

Antagonist studies have established a role for endogen- ligated, transected and reflected to allow for a midlineous CRF or CRF-like peptides in the expression of anxiety approach to DRN with minimal blood loss. The coordi-and the modulation of anxiety by stress factors. In general, nates for the DRN were 9 mm posterior to bregma, 0 mmantagonists for the CRF-1 subtype have not had strong lateral and 5.7 mm from the surface of the skull. A doubledirect anxiolytic effects in some standard animal models of barrel infusion pipette [1] was employed, one barrel for theanxiety such as the elevated plus-maze, but have had infusion of CRF and the second for the infusion of a smallprofound effects on the modulation of anxiety by stressors amount of the dye pontamine sky blue to aid in histologi-as measured in these tests [12,13,23]. A recent study has cal location of the pipette tip.found direct anxiolytic effect of a CRF-2 antagonist in the For recording in LS a tungsten microelectrode (3–5elevated plus maze [35]. megohm impedance, F.H. Haer) was employed. The

In light of the evidence that CRF plays an important role coordinates were 0.4 mm posterior to bregma, 0.6 mmin brain mechanisms of anxiety and stress, the previously lateral to the midline, and 5.8–6.7 mm from the surface ofcited evidence that CRF regulates 5-HT release in LS is of the skull. The electrode was advanced through the lateralconsiderable importance because both 5-HT and LS have septal region by means of a hydraulic microdrive (Trentbeen substantially implicated in anxiety and the stress Wells Inc.). Cellular activity was recorded by conventionalresponse [9,33,39,40]. In particular, the dorsolateral septal amplification and filtering systems (A-M systems) andregion appears to be especially involved in the modulation recorded on magnetic tape for off-line analysis. Spikeof anxiety. This is a region of the septum which contains a analysis was performed by commercial softwarehigh density of 5-HT receptors [2] as well as presynaptic (Datawave Discovery Package) on a PC.elements immunoreactive for 5-HT [8]. 5-HT applied Animals were assigned to five independent groups,directly to this region has a pronounced effect on LS including an artificial cerebrospinal fluid (ACSF) controlneural activity [14]. Most pertinently, the dorsolateral group and four dosage groups, 3, 10, 30 and 100 ng.septum is a region in which the release of 5-HT is Spontaneous activity in LS was recorded for at least 10profoundly affected by CRF administered to DRN [29]. min prior to the infusion of CRF. In each case CRF orThus, the effect of CRF on forebrain 5-HT release within vehicle was injected in a volume of 100 nl over a period ofthe lateral septum, could well be an important element in 1–2 min. Only one dose of CRF or ACSF was adminis-brain mechanisms associated with anxiety and stress. It tered to an individual rat. Activity was monitored for anwas of interest, therefore, to determine how the effects of additional 15 min after CRF infusion.CRF on the DRN systems projecting to LS impact uponLS activity. Accordingly, the purpose of the present study 2 .3. Histologywas to determine the effects of CRF administered directlyinto the DRN upon cellular unit activity in the lateral At the end of the experiment a small lesion was madeseptum. through the recording electrode (100mA anodal DC

current, for 10 s) to aid in identifying the electrode tip inthe LS and 50 nl of pontamine sky blue was infused

2 . Materials and methods through the second barrel of the two-barrel pipette toidentify the injection site. The brain was removed and

2 .1. Subjects placed in a 10% formol-saline solution. Frozen sectionswere taken at 40mm and photographed using the fresh

The subjects were male Sprague–Dawley albino rats sections as negatives [10].

E. Thomas et al. / Brain Research 960 (2003) 201–208 203

3 . Results activity in the 5 min pre-injection time period. Fig. 2Apresents the effect on unit activity for each of the doses. As

Fig. 1 depicts the locations of administration sites in may be seen, on the average the 3 and 10 ng doses had aDRN and recording sites in LS. Baseline rates of firing of facilitatory effect on unit activity in LS. The effect of 3 ngLS cells ranged from 0.02 to 20/s with a median rate of peaked at about 10 min whereas the effect of 10 ng was2.7/s. greater and more prolonged with a peak at about 15 min.

The results reveal that CRF administered to DRN had a The higher doses (30 and 100 ng), had either no effect orsubstantial dose-dependent effect upon unit activity in the an inhibitory effect upon unit activity in the LS. The 30 ngLS. dose had a potent and prolonged inhibitory effect. The 100

To examine the temporal characteristics of the effects of ng dose had mixed effects, but tended to show an earlyCRF, the 15 min post injection recording period was inhibition followed by a rebound excitation.divided into three 5-min time blocks. The activity in each Fig. 2B shows the overall effect of each dose over thetime block was expressed as a percentage of the baseline entire 15 min post-injection period expressed as percentage

Fig. 1. Representation of the locations of recording and injection sites. (A) Locations of recording sites in the lateral septum. The sites were clusteredaround the dorsolateral and intermediolateral septal nuclei. (B) Locations of injection sites in the dorsal raphe nucleus. The majority of the siteswere in thedorsal and ventral regions of the nucleus. The figure was adapted from Paxinos and Watson [25].

204 E. Thomas et al. / Brain Research 960 (2003) 201–208

Fig. 2. The effects of intraraphe CRF on unit activity in the lateral septum (61 S.E.M.). (A) The abscissa represents 5 min time periods afteradministration of CRF. The ordinate designates the mean spike rate expressed as the mean rate for each 5 min period over the 5 min period prior toinjection. The 3 and 10 ng doses had a facilitatory effect on unit activity in LS. The effect of 3 ng peaked at about 10 min whereas the effect of 10 ng wasmore potent and prolonged and peaked at about 15 min. The 30 ng dose had a potent and prolonged inhibitory effect. The 100 ng dose had mixed effects,but tended to show an early inhibition followed by a rebound excitation. (B) The bars represent the total activity in LS for the 15 min post-injection periodfor each dose of CRF or vehicle administered to DRN. The asterisks indicate a significant differnce from baseline by Student’st-test (P,0.05j.

change from the baseline activity. A one-way ANOVA dose was significantly lower than baseline (t[14]52.19,revealed a significant effect of dose,F(4,49)54.27, P5 P,0.05, two-tailed).0.005. The 10 ng dose was significantly greater than While the data for each dose yielded effects that werebaseline (t[10]52.25, P,0.05, two-tailed) and the 30 ng consistent enough to be represented as averages in the

E. Thomas et al. / Brain Research 960 (2003) 201–208 205

Table 1 responding in the LS corresponds closely to the timeThe response of units in the lateral septum to doses of CRF in the dorsal course found for its effects on DRN neuronal activity andraphe

5-HT release in the LS [16,29]. Thus, importantly, thisIncrease Decrease Biphasic No Change study demonstrates that CRF acting within the DRN can

3 ng 6 1 3 impact on a forebrain target of the DRN, the LS. The10 ng 7 1 4 findings demonstrate the functional integrity of a circuit30 ng 2 9 3 that may underlie certain behavioral responses to stress.

100 ng 2 1 3 3It has been suggested that the biphasic effect of CRF on

DRN neurons reflects the existence of multiple CRFgraphs above, inspection of the data revealed some vari- receptor subtypes in the DRN that are linked to differentance within each group. Table 1 shows the number of units consequences. Two prominent CRF receptor subtypes,showing a general excitatory or inhibitory response to each CRF-R1 and CRF-R2, have been identified in the DRNdose of CRF. To qualify as an excitatory or inhibitory [4]. The ovine form of CRF that was used in this study iseffect, activity in at least two of the three 5 min blocks had more potent at the CRF-R1 receptor site than at theto be significantly greater or smaller than baseline as CRF-R2 receptor site [21]. Thus, the findings of theevaluated by at-test. A biphasic response was one in present experiment are consistent with the notion thatwhich there was an initial significant inhibition followed CRF-R1 receptors subserve an inhibitory response in theby a significant excitation. As may be seen, while there DRN whereas CRF-R2 receptors subserve an excitatorywas some variance, overall, the proportion of units that response. Confirmation of this hypothesis awaits furtherwere either excited or inhibited reflects the dose–response pharmacological characterization with agents that arefunctions in the previous figures. For the 3 and 10 ng doses selective for specific CRF receptor subtypes. Interestingly,the great proportion of units were excited. For the 30 ng excitatory effects of CRF have been observed in a sub-dose most units were inhibited, and for the 100 ng dose population of DRN neurons in vitro [22].there was a mixed response. The concept that CRF modulates LS activity via the

Fig. 3 presents a peri-event time histogram for a DRN has important implications for the understanding ofrepresentative unit in each of the 10 and 30 ng dose the neural basis of emotion and the central adaptivegroups. Both of these are typical of the time course of the responses to stress. It is of interest to speculate how theprolonged effects of the excitatory effect of the 10 ng dose modulation of LS unit activity by intraraphe CRF and itsand the inhibitory effect of the 30 ng dose. putative effect on release of serotonin in LS might impact

Finally, while there was considerable variance in on behavior. In contrast to the medial septum, which isbaseline rates of firing, there appeared to be no correlation highly responsive to stimuli from multiple modalities, thebetween the baseline rate and the percent change due to LS is not particularly responsive to neutral stimuli [31]. InCRF administration. Pearson product moment correlations the awake, freely moving animal LS units are responsive tofor all the measures reported here varied from 0.05 to 0.41. stimuli which have emotional tone, e.g. aversive con-None of the correlations approached statistical signifi- ditioned and unconditioned stimuli [36,40]. LS cells arecance. inhibited by such stimuli, and are activated by stimuli

associated with the relief of stress. This biphasic responsemay well be modulated by the DRN serotonin system.

4 . Discussion Indeed, activation of the LS has been implicated in thereduction of the physiological response to stress [39,40],

In the present study, local injection of low doses of CRF increases in punished responding [41] and increases ininto the DRN increased unit discharge in the dorsolateral swimming in the forced swim test [5]. Recent evidenceregion of LS and this effect reversed as the dose was suggests that CRF regulation of 5-HT release in the LSincreased. Given the prior evidence that 5-HT inhibits LS plays a role in behavioral responses to swim stress. Uponneurons [15,32], these findings were remarkably consistent an initial exposure to swim stress, which is associated withwith the effects of CRF on DRN neuronal discharge [16] both active escape and passive (immobility) behaviors,and 5-HT release in the LS [29]. Thus, at low doses, 5-HT release in the LS was decreased [17,28], an effectintra-DRN administration of CRF is associated with de- that should result in LS activation. This was prevented bycreases in DRN discharge [16] and 5-HT release in the LS prior administration of a CRF antagonist, indicating that[29]. The resulting loss of a serotonergic inhibitory in- endogenous CRF mediated the neurochemical responsefluence would be predicted to activate the LS, as was [28]. In contrast, subsequent exposure to swim stress wasobserved. At higher doses (e.g., 100 ng) CRF-induced associated with a loss in the ability of either swim stress orincreases in DRN discharge would be predicted to facili- CRF to decrease 5-HT release in the LS, as well as atate the serotonergic inhibitory influence in the LS. Con- decrease in active escape behaviors [17,28]. Based on this,sistent with such an interpretation is the finding that the it is tempting to speculate that the behavioral impact oftime course of the effect of intra-DRN CRF on unit CRF on DRN-LS pathway is to facilitate active behaviors

206 E. Thomas et al. / Brain Research 960 (2003) 201–208

Fig. 3. Examples of peri-event time histograms (5 s time bins) from individual neurons for the 10 and 30 ng doses. CRF injection is indicated by the arrow.(A) illustrates the excitatory effect of the 10 ng dose upon firing rate and (B) illustrates the inhibitory effect of the 30 ng dose.

in the initial response to stress. This may serve to balance plied by Dr Jean Riviere of the Clayton Foundation forthe impact of other circuits that mediate passive coping in Peptide Biology, The Salk Institute, La Jolla, CA.response to stress.

R eferences

A cknowledgements[1] H. Akaoka, C.F. Saunier, K. Chergui, P. Charlety, M. Buda, G.

Chouvet, Combining in vivo volume-controlled pressure microejec-This study was supported by PHS grants MH54674, tion with extracellular unit recording, J. Neurosci. Methods 42

MH02006, MH58250. Ovine CRF was generously sup- (1992) 119–128.

E. Thomas et al. / Brain Research 960 (2003) 201–208 207

[2] A. Biegon, T.C. Rainbow, B.S. McEwen, Quantitative autoradiog- M. Davis, Corticotropin-releasing factor: long-lasting facilitation ofraphy of serotonin receptors in the rat brain, Brain Res. 242 (1982) the acoustic startle reflex, J. Neurosci. 12 (1992) 2303–2312.197–204. [21] T.W. Lovenberg, C.W. Liaw, D.E. Grigoriadis, W. Clevenger, D.T.

[3] K.T. Britton, Y. Akwa, M.G. Spina, G.F. Koob, Neuropeptide Y Chalmers, E.B. De Souza, T. Oltersdorf, Cloning and characteriza-blocks anxiogenic-like behavioral action of corticotropin-releasing tion of a functionally distinct corticotropin-releasing factor receptorfactor in an operant conflict test and elevated plus maze, Peptides 21 subtype from rat brain, Proc. Natl. Acad. Sci. USA 92 (1995)(2000) 37–44. 836–840, [published erratum appears in Proc. Natl. Acad. Sci. USA

92(12) (1995) 5759].[4] D.T. Chalmers, T.W. Lovenberg, E.B. De Souza, Localization ofnovel corticotropin-releasing factor receptor (CRF2) mRNA expres- [22] C.A. Lowry, J.E. Rodda, S.L. Lightman, C.D. Ingram, Corticot-sion to specific subcortical nuclei in rat brain: comparison with ropin-releasing factor increases in vitro firing rates of serotonergicCRF1 receptor mRNA expression, J. Neurosci. 15 (1995) 6340– neurons in the rat dorsal raphe nucleus: evidence for activation of a6350. topographically organized mesolimbocortical serotonergic system, J.

Neurosci. 20 (2000) 7728–7736.[5] C.M. Contreras, M. Molina, M. Saavedra, L. Martinez-Mota, Lateralseptal neuronal firing rate increases during proestrus-estrus in the [23] F. Menzaghi, R.L. Howard, S.C. Heinrichs, W. Vale, J. Rivier, G.F.rat, Physiol. Behav. 68 (2000) 279–284. Koob, Characterization of a novel and potent corticotropin-releasing

factor antagonist in rats, J. Pharmacol. Exp. Ther. 269 (1994)[6] S.F. De Boer, J.L. Katz, R.J. Valentino, Common mechanisms564–572.underlying the proconflict effects of corticotropin-releasing factor, a

benzodiazepine inverse agonist and electric foot-shock, J. Phar- [24] H. Monnikes, B.G. Schmidt, Y. Tache, Psychological stress-inducedmacol. Exp. Ther. 262 (1992) 335–342. accelerated colonic transit in rats involves hypothalamic corticot-

ropin-releasing factor, Gastroenterology 104 (1993) 716–723.[7] E.B. De Souza, T.R. Insel, M.H. Perrin, J. Rivier, W.W. Vale, M.J.Kuhar, Corticotropin-releasing factor receptors are widely distribut- [25] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates,ed within the rat central nervous system: an autoradiographic study, 4th Edition, Academic Press, New York, 1998.J. Neurosci. 5 (1985) 3189–3203. [26] E. Potter, S. Sutton, C. Donaldson, R. Chen, M. Perrin, K. Lewis,

[8] C. Gall, R.Y. Moore, Distribution of enkephalin, substance P, P.E. Sawchenko, W. Vale, Distribution of corticotropin-releasingtyrosine hydroxylase, and 5-hydroxytryptamine immunoreactivity in factor receptor mRNA expression in the rat brain and pituitary, Proc.the septal region of the rat, J. Comp. Neurol. 225 (1984) 212–227. Natl. Acad. Sci. USA 91 (1994) 8777–8781.

[9] J.A. Gray, The Neuropsychology of Anxiety: An Enquiry Into the [27] M. Price, A. Curtis, L. Kirby, R. Valentino, I. Lucki, Effects ofFunctions of the Septo-hippocampal System, Oxford University corticotropin-releasing factor on brain serotonergic activity, Neuro-Press, Oxford, 1982. psychopharmacology 18 (1998) 492–501.

[10] C.M. Guzman-Flores, M. Alcarez, A. Fernandez-Guardiola, in: [28] M.L. Price, L.G. Kirby, R.J. Valentino, I. Lucki, Evidence forRapid Procedure To Localize Electrodes in Experimental Neuro- corticotropin-releasing factor regulation of serotonin in the lateralphysiology, Vol. 16, Boletin del Instituto des Estudios Medicos y septum during acute swim stress: adaptation produced by repeatedBiologicos, Universidad Nacional de Mexico, 1958, pp. 26–31. swimming, Psychopharmacology 162 (2002) 406–414.

[11] S.E. Hammack, K.J. Richey, M.J. Schmid, M.L. LoPresti, L.R. [29] M.L. Price, I. Lucki, Regulation of serotonin release in the lateralWatkins, S.F. Maier, The role of corticotropin-releasing hormone in septum and striatum by corticotropin-releasing factor, J. Neurosci.the dorsal raphe nucleus in mediating the behavioral consequences 21 (2001) 2833–2841.of uncontrollable stress, J. Neurosci. 22 (2002) 1020–1026. [30] M. Sakanaka, T. Shibasaki, K. Lederis, Corticotropin releasing

[12] S.C. Heinrichs, F. Menzaghi, E.M. Pich, H.A. Baldwin, S. Rassnick, factor-like immunoreactivity in the rat brain as revealed by aK.T. Britton, G.F. Koob, Anti-stress action of a corticotropin- modified cobalt-glucose oxidase-diaminobenzidine method, J.releasing factor antagonist on behavioral reactivity to stressors of Comp. Neurol. 260 (1987) 256–298.varying type and intensity, Neuropsychopharmacology 11 (1994) [31] M. Segal, Convergence of sensory input on units in the hippocampal179–186. system of the rat, J. Comp. Physiol. Psychol. 87 (1974) 91–99.

[13] S.C. Heinrichs, E.M. Pich, K.A. Miczek, K.T. Britton, G.F. Koob, [32] M. Segal, Responses of septal nuclei neurons to microiontophoreti-Corticotropin-releasing factor antagonist reduces emotionality in cally administered putative neurotransmitters, Life Sci. 14 (1974)socially defeated rats via direct neurotropic action, Brain Res. 581 1345–1351.(1992) 190–197. [33] P.D. Sparks, J.E. LeDoux, Septal lesions potentiate freezing behavior

[14] M. Joels, P. Shinnick-Gallagher, J.P. Gallagher, Effect of serotonin to contextual but not to phasic conditioned stimuli in rats, Behav.and serotonin analogues on passive membrane properties of lateral Neurosci. 109 (1995) 184–188.septal neurons in vitro, Brain Res. 417 (1987) 99–107. [34] L.W. Swanson, P.E. Sawchenko, J. Rivier, W.W.Vale, Organization of

[15] M. Joels, I.J. Urban, Monoamine-induced responses in lateral septal ovine corticotropin-releasing factor immunoreactive cells and fibersneurons: influence of iontophoretically applied vasopressin, Brain in the rat brain: an immunohistochemical study, Neuroendocrinolo-Res. 344 (1985) 120–126. gy 36 (1983) 165–186.

[16] L. Kirby, K. Rice, R. Valentino, Effects of corticotropin-releasing [35] L.K. Takahashi, S.P. Ho, V. Livanov, N. Graciani, S.P. Arneric,factor on neuronal activity in the serotonergic dorsal raphe nucleus, Antagonism of CRF(2) receptors produces anxiolytic behavior inNeuropsychopharmacology 22 (2000) 148–162. animal models of anxiety, Brain Res. 902 (2001) 135–142.

[17] L.G. Kirby, I. Lucki, Interaction between the forced swimming test [36] E. Thomas, Forebrain mechanisms in the relief of fear: The role ofand fluoxetine treatment on extracellular 5-hydroxytryptamine and the lateral septum, Psychobiology 16 (1988) 36–44.5-hydroxyindoleacetic acid in the rat, J. Pharmacol. Exp. Ther. 282 [37] R.J. Valentino, S.L. Foote, M.E. Page, The locus coeruleus as a site(1997) 967–976. for integrating corticotropin releasing factor and noradrenergic

[18] G.F. Koob, S.C. Heinrichs, A role for corticotropin releasing factor mediation of stress responses, Ann. N Y Acad. Sci. 697 (1993)and urocortin in behavioral responses to stressors, Brain Res. 848 173–188.(1999) 141–152. [38] R.J. Valentino, E. Van Bockstaele, Cortocotropin-releasing factor:

[19] K.C. Liang, E.H.Y. Lee, Intra-amygdala injections of corticotropin- Putative neurotransmitter actions of a neurohormone, in: D.W. Pfaff,releasing factor facilitate inhibitory avoidance learning and reduce A.P. Arnold, A.M. Etgen, S.E. Fahrbach, R.T. Rubin (Eds.),exploratory behavior in rats, Psychopharmacology 96 (1988) 232– Hormones, Brain and Behavior, Vol. 4, Academic Press, San Diego,236. 2002, pp. 81–102.

[20] K.C. Liang, K.R. Melia, M.J. Miserendino, W.A. Falls, S. Campeau, [39] E. Yadin, E. Thomas, Stimulation of the lateral septum attenuates

208 E. Thomas et al. / Brain Research 960 (2003) 201–208

immobilization-induced stress ulcers, Physiol. Behav. 59 (1996) Frontiers in Stress Research: Modulation of Brain Function, Har-883–886. wood Academic Publishers, Amsterdam, 1998, pp. 73–81.

[40] E. Yadin, E. Thomas, Limbic mechanisms of anxiety and stress, in: [41] E. Yadin, E. Thomas, H.L. Grishkat, C.E. Strickland, The role of theA. Levy, E. Grauer, D. Ben-Nathan, R.E. de Kloet (Eds.), New lateral septum in anxiolysis, Physiol. Behav. 53 (1993) 1077–1083.