Effects of Pavlovian fear conditioning on septohippocampal metabolism in rats

Neuroscience Letters 373 (2005) 94–98

Effects of Pavlovian fear conditioning on septohippocampal metabolism in rats N´elida M. Conejoa,∗ , Mat´ıas L´opezb , Ra´ul Cantorab , H´ector Gonz´alez-Pardoa , Laudino L´opeza , Azucena Begegaa , Guillermo Vallejoc , Jorge L. Ariasa a

Psychobiology Laboratory, Department of Psychobiology, Facultad de Psicolog´ıa, Plaza Feijoo, s/n E-33003 Oviedo, Spain b Animal Learning Laboratory, Department of Psychology, University of Oviedo, Oviedo, Spain c Methodology Area, Department of Psychology, University of Oviedo, Oviedo, Spain Received 22 July 2004; received in revised form 17 September 2004; accepted 28 September 2004

Abstract The effects of classical fear conditioning in different regions of the limbic system were analysed using cytochrome oxidase (CO) histochemistry. Wistar rats were submitted to different conditions. Rats in the group Paired received tone–shock pairing, to elicit conditioned suppression of lever pressing (i.e., tone will evoke conditioned fear responses). The group Unpaired underwent random presentations of these stimuli and developed no conditioned fear. Untrained animals were also included as a control group. A significant decrease in CO activity was found in the medial septal area and the dorsal hippocampus (CA3 subfield and dentate gyrus) in the group Paired as compared with the group Unpaired. Furthermore there was greater metabolic activity in the control group as compared with the other two groups. No differences in CO labelling of the basolateral amygdala were detected among all groups. These findings suggest that the septohippocampal system plays an important role in controlling conditioned fear behaviour. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Conditioned fear; Cytochrome oxidase; Septohippocampal system; Amygdala; Rat

Classical fear conditioning is a form of associative learning in which subjects come to express fear responses to a neutral conditioned stimulus (CS) that is paired with an aversive unconditioned stimulus (US). As a result of this pairing, the CS acquires the ability to elicit behavioural, autonomic, and endocrine responses that are characteristically expressed in the presence of danger. Because fear conditioning is rapidly acquired and persistent, involves well-defined stimuli and responses, occurs widely in the animal kingdom, and implicates similar neural circuits in different vertebrate species, it has emerged as an especially useful behavioural model for investigating the neurobiological mechanisms of learning and memory [11,12]. The septohippocampal system and basolateral amygdala have been involved in classical conditioning, playing a mod∗

Corresponding author. Tel.: +34 98 5103212; fax: +34 98 5104144. E-mail address: [email protected] (N.M. Conejo).

0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.09.066

ulatory role in the acquisition and retention of conditioned responses [3,12,16]. However, the functional involvement of these structures in the acquisition and expression of conditioned fear responses to aversive stimuli is not yet fully understood [2,17]. The present study used cytochrome oxidase (CO) histochemistry to evaluate oxidative metabolism, since a tight coupling exists between oxidative energy metabolism and neuronal activity [21]. CO activity is dependent on the energetic demand produced by sustained neuronal activity in the brain, and it has been applied to study regional changes in brain activity associated with classical conditioning [1,4,14]. Training involved the CER or conditioned fear paradigm [5]. In this protocol, training establishes a stimulus as predictive of an aversive event and the degree of conditioning is assessed by measuring disruption of ongoing behaviour such as leverpressing. CO histochemistry was used to evaluate the metabolic capacity of several limbic regions after classical fear conditioning.

N.M. Conejo et al. / Neuroscience Letters 373 (2005) 94–98

We used 30 male Wistar rats with free-feeding weights within the range of 364–573 g at the beginning of the experiment. The rats were housed individually in standard plastic cages (27 cm × 27 cm × 15 cm) with water freely available in a temperature (23 ◦ C) and light (12-h light:12-h dark cycle) controlled room. Throughout the experiment they were maintained on a 22.5-h schedule of food deprivation. All procedures were performed in accordance with guidelines of the European Council Directive (86/609/EEC). Behavioural procedures were performed in four, identical, standard operant chambers (26.5 cm × 22 cm × 20 cm, Letica Instruments, Spain), housed in sound-attenuating boxes. Each chamber was equipped with a single lever located to the left-hand side of a central, recessed magazine that provided access, via a flap door, to a pellet dispenser, which could deliver 45-mg food pellets (Noyes, improved Formula A). The floor of the chambers consisted of 16 stainless steel rods, 5 mm in diameter and spaced 1.5 cm, which could be electrified through a scrambler from a shock source. The auditory CS was a tone of 120 Hz, 30 s in duration, with an intensity of 75 dB, delivered through a speaker mounted on the front wall of the chambers. The US was a mild foot shock of 0.5 mA, 0.5 s in duration, delivered through the grid floor of the chambers. An IBM microcomputer controlled the equipment and recorded the lever presses during instrumental training and testing. Before training, subjects were randomly assigned to one of three groups: Paired (n = 11), Unpaired (n = 9), and Untreated control (n = 10). The first two groups underwent behavioural training which consisted of three stages: lever press training, preexposure to the tone CS, and finally either tone–shock pairings (group Paired) or random presentations of the stimuli (group Unpaired). Rats assigned to the Untreated control group lacked behavioural training. Initially, the rats were trained to collect food rewards (45 mg Noyes food pellets) during two, 30-min magazine training sessions. The rewards were delivered on a random time (RT) 60-s schedule with the levers retracted. On the next day, with the levers replaced, all the rats were trained to respond on the lever with a continuous schedule of reinforcement (i.e., each lever press was rewarded). The session continued until the rat had obtained 30 rewards or 1 h had elapsed. If animals did not complete 30 lever presses in an hour, they underwent an additional training session with the continuous schedule of reinforcement. In the next four sessions of lever training rewards were delivered according to a variable interval (VI) schedule (i.e., on average the rat’s first response after each interval is reinforced), whose time parameter was increased from 5 to 15 s and 30 to 60 s across successive sessions. Sessions lasted 40 min. The VI-60 s schedule was maintained until the completion of the experiment. Upon the completion of the lever press training sessions, the animals in group Paired and group Unpaired received preexposure to the 120 Hz tone. Preexposure comprised a single 40 min session during which there were two 30 s pre-

95

sentations of the tone. The first tone presentation was accomplished 10 min after the start of the session and the second tone presentation 20 min later. No shock was delivered during this session. This session served to remove the initial disruption of lever pressing observed following presentation of any novel stimulus. Following the preexposure session, there were 8 days of classical conditioning for the group Paired. They consisted of three pairings a day of the tone, again of 30-s duration, with the footshock for a total of 24 paired presentations. The tone terminated with the footshock. Each session was 40 min long, with 10 min between each tone–shock pairing. Eight minutes elapsed from the start of the session to the onset of the first tone–shock presentation. For the group Unpaired, daily random training consisted of alternating presentations of three tones and three shocks over 40 min, with 5 min between each presentation. The total numbers of tone and shock stimuli were the same in both groups. The degree of fear conditioning was assessed by determining if the normally ongoing lever press response is disrupted by the tone presentation. Suppression to the tone was measured by the ratio A/(A + B), where A represents the number of lever presses made during the 30-s presentation of the stimulus and B the number of lever presses made during the 30-s immediately prior to the onset of the stimulus (preCS scores). Hence, a ratio of 0.50 represents no suppression during the stimulus, and a ratio of 0.00 represents maximal conditioned suppression. Evidence of fear conditioning was evaluated by comparing the conditioned suppression ratios to tone across aversive conditioning sessions using a repeated measures two-way ANOVA (group × session block). The performance during the lever press training and the unconditioned suppression during the preexposure session were analysed with ANOVA as well. One hour following the onset of last aversive conditioning session, animals were sacrificed by overdose of sodium pentobarbital (100 mg/kg) and changes in the metabolic activity of selected brain regions were analysed using quantitative cytochrome oxidase histochemistry, following the method by Wong-Riley [20]. Briefly, 20 ␮m-thick brain sections were obtained using a cryostat microtome (Microm, Heidelberg) and incubated in darkness for 2 h at 37 ◦ C in a staining bath containing cytochrome c (Sigma, USA), sucrose, and diaminobenzidine tetrahydrochloride (Sigma) in 0.1 M phosphate buffer (pH 7.4). Finally, the sections were rinsed in phosphate buffer, dehydrated in alcohol and coverslipped with Entellan (Merck, Germany). A series of sections of rat liver cut at different thicknesses (10, 20, 40, and 80 ␮m) were included together with brain tissue in each bath. These sections were used as standards to control for staining variability across different incubation baths, as previously described [8]. Using an image processing system (Leica Q-550, Germany), relative optical density (OD) readings were obtained from the dorsal hippocampus (CA1, CA3 and dentate gyrus), medial septum and basolateral amygdala in each subject (six readings taken bilaterally in each brain region). Optical den-

96

N.M. Conejo et al. / Neuroscience Letters 373 (2005) 94–98 Table 1 Effects of classical fear conditioning on CO activity Limbic structures

Group Paired Group Unpaired Untreated group

CA1 subfield CA3 subfield Dentate gyrus Medial septum Basolateral amygdala

48.7 53.3 59.2 63.1 71.9

± ± ± ± ±

3.3 2.6 2.8 2.4 3.7

56.1 61.7 70.0 72.7 70.4

± ± ± ± ±

2.5 3.2* 3.0* 4.1* 1.9

68.67 72.82 81.38 84.03 71.53

± ± ± ± ±

3.34**† 2.79*† 1.69**† 2.40*† 4.17

Data represent mean (±S.E.M.) relative optical density. Significant differences vs. Group Paired (Student’s t-test, * p < 0.05 and ** p < 0.01, two-tailed). Significant differences vs. Group Unpaired (Student’s t-test, † p < 0.05, twotailed). Fig. 1. Mean (±S.E.M.) suppression ratios to tone during aversive conditioning sessions in two-session blocks. A ratio of 0.50 represents no conditioned fear during the tone, and a ratio of 0.00 represents maximal fear conditioning. Significant group differences (Student’s t-test, p < 0.01).

sity readings between baths were corrected by comparing the measures taken from each liver standard. Differences in CO staining (OD) of each brain region between the selected groups were analyzed using one way ANOVA. Post hoc comparisons were performed by Tukey’s tests. Pearson’s moment–product correlation coefficients were calculated between mean OD measured in each brain region and mean suppression ratio for each subject of the group Paired. Values are expressed as mean ± S.E.M. and p < 0.05 was considered significant in this study. The performance during the last session of lever press training did not differ between the Paired and Unpaired groups (F < 1). The mean (±S.E.M.) number of lever presses during this session for the two groups was: Paired, 766 ± 104; Unpaired, 660 ± 64. The differences between the unconditioned suppression ratios to the tone during the preexposure session were not significant [F(1,18) = 1.60; p = 0.22]. The mean (±S.E.M.) unconditioned suppression ratio for each group during this session was: Paired, 0.41 ± 0.03; Unpaired, 0.46 ± 0.02. The mean suppression ratios to tone for the two groups during classical conditioning phase are showed in the Fig. 1 in blocks of two sessions. The lever press rate in group Paired decreased across sessions, indicating acquisition of classical fear conditioning to tone. In contrast, there was no evidence of conditioned suppression to the tone in group Unpaired. The analysis revealed significant effects of group [F(1,18) = 149.79; p < 0.001] and session block factors [F(3,54) = 25.43; p < 0.001], together with a group × block interaction [F(3,54) = 4.32; p = 0.008]. This pattern of results is exactly that expected on the assumption that the group Paired did learn to suppress lever press behaviour during tone presentations (i.e., tone evoked conditioned fear responses), whereas the group Unpaired did not suppress. Significant decreases in CO activity were found in both the Paired and Unpaired groups as compared to the Untreated group, in the following regions: the medial septal area [F(2,28) = 13.48; p < 0.001], the dorsal region of CA1 [F(2,28) = 10.73; p < 0.001] and CA3 sub-

fields [F(2,28) = 12.94; p < 0.001], and the dentate gyrus [F(2,28) = 20.36; p < 0.001] of the hippocampus. However, no significant group differences were found between Paired and Unpaired groups in the CA1 subfield. In addition, no differences were found among all groups in the basolateral amygdala (see Table 1). Furthermore, neurobehavioural correlations between the conditioned suppression ratio on the last conditioning session and the metabolic capacity were found in the medial septum (r = 0.84, p = 0.001; Fig. 2a) and the dorsal region of CA3 subfield (r = 0.71, p = 0.014; Fig. 2b) in the group Paired (n = 11). To our knowledge, this is the first time that CO histochemistry is used to study changes in limbic regions induced by conditioned fear in the rat. A previous study using CO histochemistry to evaluate the effects of classical conditioning on neuronal activity was particularly focused on the auditory system [14]. CO activity is different from other markers of brain metabolism, because it reflects long-term cumulative changes in metabolic activity resulting from the sustained

Fig. 2. (a) and (b) Neurobehavioural correlations between CO activity and mean suppression ratio in the medial septum (a) and CA3 hippocampal subfield (b). a.u.: arbitrary units.

N.M. Conejo et al. / Neuroscience Letters 373 (2005) 94–98

increase in the energy demands of brain cells in vertebrates [15]. In mammals, CO histochemistry reveals the cumulative changes in brain activity associated with the entire learning period, but it is not an index of ongoing evoked activity as compared with other functional brain mapping techniques [15]. Our results showed a lower CO activity of the septohippocampal system in the trained groups as compared to the untrained group. This result could be interpreted as a general involvement of this system in fear and anxiety-inducing situations, as previously hypothesized [9]. In addition, decreases of CO activity in these regions in the group Paired as compared with the group Unpaired, are consistent with the view that neural changes associated with auditory fear conditioning depend on the learned signal value of the tone. In agreement with our data, other authors reported a reduced activity of the same regions in rats showing conditioned response inhibition [10]. The hippocampus is considered to mediate the acquisition and consolidation of a memory for the aversive conditioning context [6,18]. In particular, lesions of the dorsal hippocampus impair classical fear conditioning in rats [13]. In addition, medial septum lesions impair classical conditioning in rabbits and humans [16]. The lack of significant differences in the CO activity of the basolateral amygdala among all groups could be interpreted as a time-limited involvement of this brain region when a fearinducing stimulus is present. In this regard, the basolateral amygdala could be activated mainly during the acquisition of the conditioned response, because it is necessary for the association between the CS and US [12], returning to its basal activity levels after memory consolidation. On the other hand, positive neurobehavioural correlations were found in group Paired between CO activities measured in the medial septum and CA3 hippocampal subfield and the mean suppression ratios. This result would support the involvement of these brain regions in the conditioned inhibition of behaviour [10,19]. The bidirectional communication between septum and dorsal hippocampus may mediate memory consolidation during associative learning [7,16,22]. In conclusion, our results shown that the medial septum and particular regions of the dorsal hippocampus are specifically involved in classical fear conditioning in rats. These findings are supported by other studies using different methods, and confirm the relevance of the septohippocampal system in fear conditioning. Moreover, they show that cytochrome oxidase histochemistry can be successfully used to assess regional long-lasting effects of learning throughout the rat brain.

Acknowledgements We wish to thank the laboratory technicians Piedad Burgos and Bego˜na Vald´es for their help. This research was sup-

97

ported by Grant MCYT BSO-2002-03404 (Ministry of Science and Technology, Spain) to Mat´ıas L´opez and by Grant MCYT BSO-2001-2757 (Ministry of Science and Technology, Spain) to Jorge Arias.

References [1] V. Agin, R. Chichery, M.-P. Chichery, Effects of learning on cytochrome oxidase activity in cuttlefish brain, NeuroReport 12 (2001) 113–116. [2] H.T. Blair, G.E. Schafe, E.P. Bauer, S.M. Rodrigues, J.E. LeDoux, Synaptic plasticity in the lateral amygdala: a cellular hypothesis of fear conditioning, Learn. Mem. 8 (2001) 229–242. [3] K.M. Christian, R.F. Thompson, Neural substrates of eyeblink conditioning: acquisition and retention, Learn. Mem. 10 (2003) 427–455. [4] P. Deglise, M. Dacher, E. Dion, M. Gauthier, C. Armengaud, Regional brain variations of cytochrome oxidase staining during olfactory learning in the honeybee (Apis mellifera), Behav. Neurosci. 117 (2003) 540–547. [5] W.K. Estes, B.F. Skinner, Some quantitative properties of anxiety, J. Exp. Psychol. 29 (1941) 390–400. [6] M. Fendt, M.S. Fanselow, The neuroanatomical and neurochemical basis of conditioned fear, Neurosci. Biobehav. Rev. 23 (1999) 743–760. [7] A.G. Foley, L.C.B. Rønn, K.J. Murphy, C.M. Regan, Distribution of polysialylated neural cell adhesion molecule in rat septal nuclei and septohippocampal pathway: transient increase of polysialylated interneurons in the subtriangular septal zone during memory consolidation, J. Neurosci. Res. 74 (2003) 807–817. [8] F. Gonzalez-Lima, A. Cada, Cytochrome oxidase activity in the auditory system of the mouse: a qualitative and quantitative histochemical study, Neuroscience 63 (1994) 559–578. [9] J.A. Gray, N. McNaughton, The Neuropsychology of Anxiety: an Enquiry into the Functions of the Septo-hippocampal System, Oxford University Press, Oxford, 2000. [10] D. Jones, F. Gonzalez-Lima, Associative effects of Pavlovian differential inhibition of behaviour, Eur. J. Neurosci. 14 (2001) 1915–1927. [11] J.E. LeDoux, Emotion circuits in the brain, Annu. Rev. Neurosci. 23 (2000) 155–184. [12] S. Maren, Neurobiology of Pavlovian fear conditioning, Annu. Rev. Neurosci. 24 (2001) 897–931. [13] S. Maren, G. Aharonov, M.S. Fanselow, Neurotoxic lesions of the dorsal hippocampus and Pavlovian fear conditioning in rats, Behav. Brain Res. 88 (1997) 261–274. [14] A. Poremba, D. Jones, F. Gonzalez-Lima, Classical conditioning modifies cytochrome oxidase activity in the auditory system, Eur. J. Neurosci. 10 (1998) 3035–3043. [15] A. Poremba, D. Jones, F. Gonzalez-Lima, Functional mapping of learning-related metabolic activity with quantitative cytochrome oxidase histochemistry, in: F. Gonzalez- Lima (Ed.), Cytochrome Oxidase in Neuronal Metabolism and Alzheimer’s Disease, Plenum Press, New York, 1998, pp. 109–144. [16] B. Rokers, E. Mercado, M.T. Allen, C.E. Myers, M.A. Gluck, A connectionist model of septohippocampal dynamics during conditioning: closing the loop, Behav. Neurosci. 116 (2002) 48–62. [17] P. Sah, E.S.L. Faber, M. Lopez de Armentia, J. Power, The amygdaloid complex: anatomy and physiology, Physiol. Rev. 83 (2003) 803–834. [18] M.J. Sanders, B.J. Wiltgen, M.S. Fanselow, The place of the hippocampus in fear conditioning, Eur. J. Pharmacol. 463 (2003) 217–223. [19] A.Z. Weitemier, A.E. Ryabinin, Subregion-specific differences in hippocampal activity between Delay and Trace fear condition-

98

N.M. Conejo et al. / Neuroscience Letters 373 (2005) 94–98

ing: an immunohistochemical analysis, Brain Res. 995 (2004) 55– 65. [20] M.T.T. Wong-Riley, Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry, Brain Res. 171 (1979) 11–28.

[21] M.T.T. Wong-Riley, Cytochrome oxidase: an endogenous metabolic marker for neuronal activity, Trends Neurosci. 12 (1989) 94–101. [22] M. Wu, M. Shanabrough, C. Leranth, M. Alreja, Cholinergic excitation of septohippocampal GABA but not cholinergic neurons: implications for learning and memory, J. Neurosci. 20 (2000) 3900–3908.