- Email: [email protected]
Corticosterone attenuates conditioned fear responses and potentiates the expression of GABA-A receptor alpha-2 subunits in the brain structures of rats selected for high anxiety
Behavioural Brain Research 235 (2012) 30–35
Contents lists available at SciVerse ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communication
Corticosterone attenuates conditioned fear responses and potentiates the expression of GABA-A receptor alpha-2 subunits in the brain structures of rats selected for high anxiety b ´ A. Wisłowska-Stanek a,∗ , M. Lehner b , A. Skórzewska b , P. Maciejak a,b , J. Szyndler a , D. Turzynska , b a,b ´ A. Sobolewska , A. Płaznik a b
Department of Experimental and Clinical Pharmacology, Medical University of Warsaw, 26/28 Krakowskie Przedmie´scie Street, 00-927 Warsaw, Poland Department of Neurochemistry, Institute of Psychiatry and Neurology, 9 Sobieskiego Street, 02-957 Warsaw, Poland
h i g h l i g h t s Corticosterone decreased expression of conditioned fear in high anxiety rats (HR). Fear increased expression of alpha-2 subunits of GABA-A receptor in limbic structures. Corticosterone potentiated effects of conditioned fear on alpha-2 subunit expression.
a r t i c l e
i n f o
Article history: Received 30 May 2012 Received in revised form 6 July 2012 Accepted 10 July 2012 Available online 20 July 2012 Keywords: GABA-A receptor alpha-2 subunits Immunocytochemistry Brain structures Corticosterone Conditioned fear Individual differences
a b s t r a c t The aim of the experiment was to assess the effects of an acutely administered corticosterone on the expression of GABA-A receptor alpha-2 subunits in the brain structures of high (HR) and low (LR) anxiety rats (divided according to their conditioned fear-induced freezing response) subjected to a second conditioned fear session (1 week after fear conditioning). We found that corticosterone (20 mg/kg, sc) given to rats prior to the second conditioned fear session significantly enhanced a decrease in fear expression in the HR group. The behavioural effect of fear was accompanied by the increased expression of alpha-2 subunits in the basolateral amygdala (BLA) and the dentate gyrus of the hippocampus (DG) of the HR group. Corticosterone potentiated the effect of fear on alpha-2 subunit expression in the BLA, DG, the cingulate cortex area 1 and the secondary motor cortex (areas Cg1 and M2). The current study provides insight into the mechanisms that may be responsible for the beneficial effects of glucocorticoids in the therapy of some anxiety disorders. © 2012 Elsevier B.V. All rights reserved.
Acute glucocorticosteroid administration facilitates active behavioural coping in threatening situations [1]. We previously found that a single dose of corticosterone given to rats before training in a conditioned fear test significantly attenuated the freezing response when examined 24 h later [2,3]. Similarly, the acute administration of corticosterone before a session in an elevated plus maze produced an anxiolytic effect in rats [4]. It is also noteworthy that cortisol (20 mg) administered orally 1 h before each extinction-based session of psychotherapy significantly reduced anxiety symptoms during exposure to a phobic situation [5]. Recently, using a model of individual differences in fear responses of rats selected according to low and high freezing responses in the contextual fear test (defined as ‘low- and high
∗ Corresponding author. Tel.: +48 22 8262116; fax: +48 22 8262116. E-mail address: [email protected] (A. Wisłowska-Stanek). 0166-4328/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2012.07.018
anxiety’ rats; LR and HR groups, respectively), we found that HR rats had deficits in the activity of the brain structures that control the cognition necessary to cope with stress (i.e., the prefrontal cortex, as measured by c-Fos expression) and increased activity of the amygdalar nuclei that enhance the stress response (c-Fos/glucocorticoid receptors-ir) [6]. We also observed that some behavioural or pharmacological interventions attenuated the increased fear responses of HR rats. For example, we found that the administration of dcycloserine and midazolam before the testing session attenuated the freezing of HR rats and increased the expression of alpha-2 subunits of the GABA-A receptor in limbic areas and the prefrontal cortex [7]. Moreover, other preclinical data suggest that a decrease in conditioned fear is correlated with the upregulation of GABAergic markers (i.e., alpha-2 subunits of the GABA-A receptor) in the amygdala [8]. The role of alpha-2 subunits in fear processing is evidenced by the finding that mice with point-mutated alpha-2 GABA-A receptor subunits are resistant to the anxiolytic-like effects
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
of benzodiazepines and display a greater behavioural response to fear-conditioned stimuli [9]. In light of these data, we decided to test a hypothesis that the corticosterone-induced decrease in rat freezing responses is related to the drug and stress-induced changes in the expression of alpha-2 subunits in limbic and cortical areas. The experiments were performed on a cohort of 71 male Wistar rats (180–200 g body weight) purchased from a licenced breeder and housed in standard laboratory conditions under a 12 h light/dark cycle (lights on at 7 a.m.). The experiments were performed in accordance with the European Communities Council Directive of November 24th, 1986 (86/609 EEC). The Local Committee for Animal Care and Use at Warsaw Medical University, Poland approved all experimental procedures using animal subjects. After 4 days of acclimatisation, 40 animals (S group) were subjected to a conditioned fear test as previously described by Wisłowska-Stanek et al. [7]. In short, the experiment was performed over 3 consecutive days. On the first day, the animals were individually placed in a training box for 2 min to adapt to the experimental conditions. The following day, after 5 min of habituation to the training box, animals underwent a fear conditioning procedure that consisted of 3 footshocks (0.7 mA, 1 s, repeated every 59 s), and the animals were removed from the testing box 3 min after the last footshock was delivered. Conditioned fear was tested on the third day (test day, T1) by re-exposing rats to the testing box and recording the freezing response over 10 min. Freezing behaviour was measured by photo beams (10 Hz detection rate) controlled by a fear conditioning PC-program. The animals were divided according to their context-induced freezing responses into LR, low-anxiety rats with total durations of freezing responses one SEM or more below the mean (i.e., <224.1 s, mean = 246 s and SEM = 21.9 s), and an HR group consisting of high-anxiety rats with total durations of freezing responses one SEM or more above the mean (i.e., >267.9 s mean = 246 s, SEM = 21.9 s). These animals were defined as the LR-T1 (n = 18) and HR-T1 (n = 16) group, respectively. The control group, C (n = 6), was not conditioned but only placed in the conditioning boxes (mean freezing duration = 92.7 s, SEM = 24.45). After testing, animals remained undisturbed in their home cages for 7 days and were then subjected to the aversive context in a second fear testing session (T2). Before the T2 session, the pre-selected groups, LR-T1 and HR-T1, were further divided into the following subgroups: LR-T2-v, low-anxiety rats given vehicle
31
(n = 7); LR-T2-cort, low-anxiety rats given corticosterone (n = 9); HR-T2-v, high-anxiety rats given vehicle (n = 8); and HR-T2-cort, high-anxiety rats given corticosterone (n = 8). The corticosterone (20 mg/kg; Sigma–Aldrich, Poland) was suspended in sesame oil and administered subcutaneously (sc) to the nape of the neck 30 min before the T2 session. Control animals were administrated sesame oil (Sigma–Aldrich, Poland). All rats were decapitated 2 h after drug or vehicle administration. A control study was performed on a second group of animals (n = 25). The rats [LR < 230.7 s (256.4 − 25.7 s); HR > 282.1 s (256.4 + 25.7 s)] were divided into four groups: LR-v, low-anxiety rats given vehicle (n = 6); LR-cort, low-anxiety rats given corticosterone (n = 5); HR-v, high-anxiety rats given vehicle (n = 6); and HR-cort, high-anxiety rats given corticosterone (n = 8). The design of the second part of the experiment was the same as the first part of the experiment with the exception that the second fear test session was not performed (T2). Thus, these animals were injected with vehicle or corticosterone 7 days after T1 and decapitated 2 h later. Immunocytochemical staining for the alpha-2 subunit of the GABA-A receptor was performed on slide-mounted frozen brain sections. Staining and counting techniques have been described previously by Wisłowska-Stanek et al. [7]. Cells were counted in the following subregions: the cingulate cortex (areas 1 and 2), the secondary motor cortex (AP: 1.20; Cg1, Cg2, M2), the basolateral amygdala and the dentate gyrus of the hippocampus (AP – 3.14; DG, BLA) [10] (Fig. 1). Three slices from each section of the brain were taken. Western blot analyses performed previously with the same antibody confirmed the specific binding of this antibody to the alpha-2 subunit. Examples of alpha-2 subunit GABA-A receptor immunopositive cells have been published by Lehner et al. [11]. The behavioural and biochemical data are presented as the means ± the SEM. The differences between the conditioned group (S) and the control group (C) were analysed by Student’s t-test. The differences in freezing response durations during the first and second conditioned fear test (T1 and T2) in the LR and HR groups were analysed by one-way repeated measures ANOVA followed by Tukey’s post hoc test. The behavioural data from the second test session were analysed by two-way ANOVA followed by Tukey’s post hoc test. The immunocytochemical data were analysed by three-way ANOVA followed by Tukey’s post hoc test. For correlation analyses, a Pearson coefficient was calculated. Statistical analyses
Fig. 1. Diagrams adapted from Paxinos and Watson [10] showing regions of the brain analysed for expression of alpha-2 subunits. BLA: basolateral amygdala; Cg1, Cg2: cingulate cortex area 1 and 2; CPu: caudate putamen; DG: dentate gyrus of the hippocampus; M2 area: prefrontal cortex, secondary motor cortex. Shaded areas indicate the analysed brain regions.
32
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
Fig. 2. Freezing behaviour during the first fear conditioning test (T1) and second fear session (T2). (A) Behaviour of saline pretreated rats. LR-T1: low-anxiety rats, first contextual fear test (n = 18); HR-T1: high-anxiety rats, first contextual fear test (n = 16); LR-T2-v: low anxiety rats, second contextual fear session (n = 7); HR-T2-v: high-anxiety rats, second contextual fear session (n = 8). (B) Freezing duration during the second fear session after pretreatment of rats with corticosterone or saline (T2). LR-T2-v: low-anxiety rats pretreated with vehicle (n = 7); HR-T2-v: high-anxiety rats, pretreated with vehicle (n = 8); LR-T2-cort: low-anxiety rats pretreated with corticosterone (n = 9); HR-T2-cort: high-anxiety rats pretreated with corticosterone (n = 8). Data are shown as the means ± SEM. *Differs from LR-T1, **p < 0.01; & differs from LR-T2-v, && p < 0.01; # differs from HR-T2-v, # p < 0.05.
were performed using Statistica 8.0 for Windows (StatSoft Inc., USA). A Student’s t-test revealed a significant difference in freezing response durations between the conditioned (S) and control (C) groups in the conditioned fear test; i.e., the effect of contextual fear conditioning; t = 2.66, df = 44, P = 0.01. ANOVA revealed significant differences in freezing durations between the first (T1) and second fear (T2) sessions and significant effects of group [F(1,15) = 65.39; p < 0.01] and time [F(1,15) = 7.79; P = 0.01], but there was no significant group × time interaction effect [F(1,15) = 0.68; P = 0.42]. Post hoc tests revealed longer freezing durations in the HR-T1 group compared to the LR-T1 group (P < 0.01) and in the HR-T2-v group compared to the LR-T2-v group (p < 0.01) (Fig. 2A). ANOVA showed significant differences in freezing durations between HR and LR rats after vehicle and corticosterone administration during the T2 session: a significant effect of group [F(1,30) = 11.59; P < 0.01], a significant group × drug interaction effect [F(1,30) = 5.14; P < 0.05] and no drug effect [F(1,30) = 3.80; P = 0.06]. Post hoc tests showed significantly longer freezing durations in the HR-T2-v group compared with the LR-T2-v group (P < 0.01) and a significant decrease
in freezing durations in the HR-T2-cort rats compared to HR-T2-v rats (P < 0.05) (Fig. 2B). ANOVA showed statistically significant differences between groups in the density of cells expressing alpha-2 GABA-A receptor subunits in Cg1, M2, BLA and DG (Table 1). Conditioned fear enhanced the expression of alpha-2 in the BLA and DG of the HRT2-v group (P < 0.01) and in the DG and Cg1 of the LR-T2-v rats (P < 0.01). The steady-state levels of alpha-2 subunits were the same in both groups; however, after the second fear session, the concentration was higher in the BLA (P < 0.01) and lower in the DG (P < 0.01) of HR-T2-v rats than LR-T2-v rats. Tukey’s post hoc test revealed higher expression of alpha-2 subunits in the HR-T2-cort group compared to the HR-T2-v and HR-cort groups in the Cg1, M2, BLA and DG (P < 0.01). Moreover, the LR-T2-cort group had higher expression of alpha-2 subunits in M2 (P < 0.01) and the DG (P < 0.05) compared to the LR-T2-v group. Post hoc analyses also revealed enhanced expression of alpha-2 subunits in Cg1, M2, DG (P < 0.01) and BLA (P < 0.05) in LR-T2-cort group compared to the LR-cort group. Correlation analyses revealed a significant negative relationship between freezing times during the T2 session and the expression of alpha-2 subunits in area M2 [r = (−)0.61; P = 0.01], the BLA [r = (−)0.70; P = 0.01] and the DG [r = (−)0.59; P = 0.02] in the HR group. In the LR group, correlation analyses did not reveal any significant effects. The main finding of this study shows that corticosterone given to rats prior to a second exposure to an aversive context significantly enhanced a spontaneously occurring decrease in fear expression in the HR group. The expression of alpha-2 subunits was increased by contextual fear in the BLA and DG of the HR group and in Cg1 and the DG of LR animals. Corticosterone potentiated the effect of fear on alpha-2 subunits in all examined structures except Cg2 in the HR group (HR-T2-cort vs. HR-T2-v). A less potent effect was present in the LR-T2-cort vs. LR-T2-v group (DG and M2 area). Importantly, pretreatment of rats with corticosterone alone did not change the expression of alpha-2 subunits in the control groups that were not exposed to an aversive context. These data implicate the role of GABAergic innervation of cortical and limbic structures in the central effects of corticosterone on fear responses. The amygdala, prefrontal cortex and hippocampus have long been recognised as a neural system responsible for the generation and elaboration of conditioned fear [cf. 12]. It is well known that this neural substrate of emotions is under tonic inhibitory control by GABAergic mechanisms [13]. The GABAergic system is not only important for the acquisition of conditioned fear but also in the attenuation of fear responses. In an experiment that evaluated training-induced changes in the expression of GABA-A-associated genes in the amygdala during the extinction of Pavlovian fear, fear extinction increased the levels of the alpha-2 and beta-2 GABA-A receptor subunit and GAD67 mRNA and decreased GABA transporter-1 mRNA [8]. These results show that fear attenuation coincides temporally with an upregulation of GABAergic markers related to enhanced GABAergic transmission. The prefrontal cortex may influence contextual fear conditioning by contributing to the integration of information about the encoded environment via the amygdala and contextual information from the hippocampus [14]. The role of the medial prefrontal cortex in this process has been documented. Van Eden et al. have shown with anterograde labelling that Fr2 (M2 area) and the anterior cingulate area (Acd) receive more projections from somatosensory and associational visual cortices than from the primary motor cortex [15]. Furthermore, the study by Hoover and Vertes [16] indicates that the information from widespread areas of the cortex is presumably integrated by the dorsal prefrontal cortex (the medial frontal agranular areas (AGm-Fr2-M2) and the anterior cingulate (ACd)) during goal directed action.
Table 1 Alpha-2 subunits of GABA-A receptor immunostaining after drug injection and the re-exposition to contextual fear test.
Cg1
HR-v (n = 6)
HR-cort (n = 8)
LR-v (n = 6)
LR-cort (n = 5)
HR-T2-v (n = 8)
HR-T2-cort (n = 8)
222.1 ± 7.6
192.1 ± 13.9
203.3 ± 6.5
203.4 ± 14.6
262.3 ± 24.2
350.8 ± 20.1bb, **
LR-T2-v (n = 7) 327.4 ± 16.6aa
LR-T2-cort (n = 9)
ANOVA
320,5 ± 18.3bb
1
[F(1,44) = 0.33 (P = 0.56)] [F(1,44) = 1.18 (P = 0.28)] 3 [F(1,44) = 86,24 (P < 0.01)] 4 [F(1,44) = 1.9 (P = 0.17)] 5 [F(1,44) = 0.79 (P = 0.37)] 6 [F(1,44) = 5,54 (P < 0.05)] 7 [F(1,44) = 7.01 (P = 0.01)] 2
235.0 ± 9.4
234.7 ± 22.2
216.6 ± 7.6
198.8 ± 10.9
216.6 ± 7.6
228.7 ± 12.5
236.3 ± 14.5
230.7 ± 12.0
1
[F(1,50) = 1.3 (P = 0.25)] [F(1,50) = 0.39 (P = 0.53)] [F(1,50) = 1.03 (P = 0.31)] 4 [F(1,50) = 0.29 (P = 0.59)] 5 [F(1,50) = 2.42 (P = 0.12)] 6 [F(1,50) = 0.07 (P = 0.78)] 7 [F(1,50) = 0.11 (P = 0.73)] 2 3
M2
145.0 ± 6.7
149.4 ± 5.8
135.0 ± 7.4
134.3 ± 6.5
181.6 ± 15.7
355.1 ± 16.8bb, **,##
143.3 ± 16.8
270.0 ± 14.9bb,&&
1
[F(1,49) = 15.09 (P < 0.01)] [F(1,49) = 63.75 (P < 0.01)] 3 [F(1,49) = 102.11 (P < 0.01)] 4 [F(1,49) = 1.74 (P = 0.19)] 5 [F(1,49) = 6.63 (P < 0.05)] 6 [F(1,49) = 59.611 (P < 0.01)] 7 [F(1,49) = 1.27 (P = 0.26)] 2
BLA
153.0 ± 14.1
168.9 ± 12.2
155.2 ± 14.1
162.8 ± 15.4
260.6 ± 12.8aa,&&
333.1 ± 11.9bb, **,##
194.00 ± 12.0
212.1 ± 11.3b
1
[F(1,49) = 26.98 (P < 0.01)] [F(1,49) = 9.59 (P < 0.01)] 3 [F(1,49) = 95.34 (P < 0.01)] 4 [F(1,49) = 2.89 (P = 0.09)] 5 [F(1,49) = 24.82 (P < 0.01)] 6 [F(1,49) = 3.32 (P = 0.07)] 7 [F(1,49) = 1.57 (P = 0.22)] 2
DG
104.0 ± 4.5
100.2 ± 3.4
91.8 ± 4.8
94.8 ± 4.6
155.9 ± 12.8aa,&&
252.1 ± 13.7bb, **
204.7 ± 12.8aa
237.7 ± 14.8bb,&
1
[F(1,49) = 0.28 (P = 0.59)] [F(1,49) = 16.76 (P < 0.01)] 3 [F(1,49) = 214.57 (P < 0.01)] 4 [F(1,49) = 3.24 (P = 0.07)] 5 [F(1,49) = 2.75 (P = 0.10)] 6 [F(1,49) = 17.17 (P < 0.01)] 7 [F(1,49) = 4.99 (P < 0.05)]
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
Cg2
2
Data are shown as means ± SEM (cell number/1 mm2 ). 1 Statistically significant group effect, 2 statistically significant drug effect, 3 statistically significant fear effect, 4 statistically significant group × drug interaction, 5 statistically significant group × fear interaction, 6 statistically significant drug × fear interaction, 7 statistically significant group × drug × fear interaction. a Differs from HR-v or LR-v, respectively, aa P < 0.01; b differs from HR-cort or LR-cort, respectively, bb P < 0.01; *differs from HR-T2-v, **P < 0.01; & differs from LR-T2-v, & P < 0.05; && P < 0.01; # differs from LR-T-cort, ## P < 0.01. For more details see section ‘Materials and methods’.
33
34
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
These findings were extended by our previous results indicating that LR animals had greater activity (as measured by c-Fos expression) of the dorsomedial part of the prefrontal cortex (area M2) than the HR group [6]. It is also noteworthy that HR and LR rats have different fear coping mechanisms. Contextual fear exposition caused less freezing and more aversive ultrasonic vocalisation in the LR group compared to HR rats [17]. The less anxious behaviour of the LR rats was not related to a higher pain threshold [18]. Moreover, our data excluded a role for deficits in cognitive processes because no differences between LR and HR rats were found in the Morris water maze (unpublished data). Recently, we also showed that the less anxious behaviour of LR animals given saline was accompanied by elevated basal levels of glutamate in the BLA and a stronger elevation of GABA in response to contextual fear when compared to HR rats. Pretreatment of rats with d-cycloserine and midazolam, drugs used in the treatment of post-traumatic stress disorder, significantly increased the concentration of GABA in the BLA and inhibited the expression of contextual fear in the HR group [19]. Moreover, these drugs caused an increase in alpha-2 subunits in limbic and cortical areas [7]. In the current study, a significant decrease in the freezing responses in the second fear session after corticosterone administration in the HR animals was accompanied by an increase in the expression of alpha-2 subunits in Cg1 and M2 of the prefrontal cortex and the DG and BLA. Thus, a drug- or corticosterone-induced increase in the expression of alpha-2 subunits in limbic structures and prefrontal cortical areas in HR rats may be a compensatory mechanism that facilitates active coping in a threatening situation. The increase in the expression of the alpha-2 subunits could be the result of enhanced receptor protein synthesis or stimulation of their active transport to the cell surface, a process that may be regulated by protein kinase C [20]. Corticosterone can significantly contribute to the rapid cycling of GABA-A receptors between the cell surface and the subsynaptic pool, which would provide a mechanism for the short-term regulation of GABAergic neurotransmission [20,21]. It has been demonstrated that glucocorticoid and mineralocorticoid receptors (GRs and MRs) are quickly translocated from the cytoplasm to the nucleus after corticosterone treatment through association with importins [21]. In the nucleus, the interaction of GRs with glucocorticoid response elements and transcriptional components and receptor clearance are dynamically regulated within seconds and minutes [22]. These studies indicate that glucocorticoids may rapidly modify the process of nuclear activation and the production of various receptor proteins and their intracellular cycling. However, the exact molecular nature of the changes in the alpha-2 subunit expression after corticosterone remains unclear, and the literature lacks a direct reference to this problem. Another rapid, non-genomic, mechanism of corticosterone action might involve the formation of GABA-A receptor-modulating neurosteroids from a corticosterone derivative, deoxycorticosterone. These derivatives, including allopregnanolone, can bind to GABA-A receptors composed of alpha-2 subunits and evoke benzodiazepine-like effects, e.g., anxiolytic-like actions [23]. GABAA receptor modulators may alter the neuroendocrine stress response, which can affect cortical control over the limbic nuclei and can also affect emotional memory [24]. In conclusion, the current study shows that animals that are more vulnerable to fear-evoking stimuli differ in the mechanisms that control GABA-A receptor alpha-2 subunit expression in prefrontal cortical areas and limbic structures. Further, our results implicate glucocorticoid-related mechanisms in this phenomenon. The current study also provides insight into the central mechanisms responsible for the beneficial effects of glucocorticoids in the therapy of some anxiety disorders. It should be also noted that similar effects have previously been found after midazolam,
d-cycloserine and corticosterone administration, indicating that their mechanisms of action in anxiety disorders are likely similar.
Acknowledgements This study was supported by statutory Grant No. 50100312043 from the Institute of Psychiatry and Neurology and by Grant No. 0440/B/P01/2009/36 from the Ministry of Science and Higher Education (ML), Warsaw, Poland.
References [1] Clay R, Hebert M, Gill G, Stapleton LA, Pridham A, Coady M, et al. Glucocorticoids are required for extinction of predator stress-induced hyperarousal. Neurobiology of Learning and Memory 2011;96:367–77. ´ ´ [2] Skórzewska A, Bidzinski A, Lehner M, Turzynska D, Wisłowska-Stanek A, Sobolewska A, et al. The effects of acute and chronic administration of corticosterone on rat behavior in two models of fear responses, plasma corticosterone concentration, and c-Fos expression in brain structures. Pharmacology Biochemistry and Behavior 2006;85:522–34. ´ ´ A, Lehner M, Turzynska D, Sobolewska A, Hamed A, [3] Skórzewska A, Bidzinski et al. The effects of acute corticosterone administration on anxiety, endogenous corticosterone, and c-Fos expression in the rat brain. Hormones and Behavior 2007;52:317–25. [4] Andreatini R, Leite JR. The effect of corticosterone in rats submitted to the elevated plus-maze and to pentylenetetrazol-induced convulsions. Progress in Neuro-Psychopharmacology and Biological Psychiatry 1994;18: 1333–47. [5] de Quervain DJ, Bentz D, Michael T, Bolt OC, Wiederhold BK, Margraf J, et al. Glucocorticoids enhance extinction-based psychotherapy. Proceedings of the National Academy of Sciences of the United States of America 2011;108:6621–5. ´ [6] Lehner M, Taracha E, Maciejak P, Szyndler J, Skórzewska A, Turzynska D, et al. Colocalisation of c-Fos and glucocorticoid receptor as well as 5-HT1A and glucocorticoid receptor immunoreactivity-expressing cells in the brain structures of low and high anxiety rats. Behavioural Brain Research 2009;200: 150–9. [7] Wisłowska-Stanek A, Lehner M, Skórzewska A, Maciejak P, Szyndler J, ´ Turzynska D, et al. Effects of d-cycloserine and midazolam on the expression of the GABA-A alpha-2 receptor subunits in brain structures of fear conditioned rats. Behavioural Brain Research 2011;225:655–9. [8] Heldt SA, Ressler KJ. Training-induced changes in the expression of GABA-A associated genes in the amygdala after acquisition and extinction of Pavlovian fear. European Journal of Neuroscience 2007;26:3631–44. [9] Morris HV, Dawson GR, Reynolds DS, Atack JR, Stephens DN. Both alpha2 and alpha3 GABAA receptor subtypes mediate the anxiolytic properties of benzodiazepeine site ligands in the conditioned emotional paradigm. European Journal of Neuroscience 2006;23:2495–504. [10] Paxinos G, Watson CH. The rat brain in stereotaxic coordinates. San Diego, CA: Academic Press Inc; 1998. [11] Lehner M, Wisłowska-Stanek A, Skórzewska A, Maciejak P, Szyndler J, ´ D, et al. Differences in the density of GABA-A receptor alpha-2 subTurzynska units and gephyrin in brain structures of rats selected for low and high anxiety in basal and fear-stimulated conditions, in a model of contextual fear conditioning. Neurobiology of Learning and Memory 2010;94:499–508. [12] Gray J. A theory of anxiety: the role of the limbic system. Encephale 1983;9:161B–6B. [13] Santos JM, Macedo CE, Brandao ML. Gabaergic mechanisms of hypothalamic nuclei in the expression of conditioned fear. Neurobiology of Learning and Memory 2008;90:560–8. [14] Hurley KM, Herbert H, Moga˛ MM, Saper CM. Efferent projection of infralimbic cortex of the rat. Journal of Comparative Neurology 1991;308:249–76. [15] Van Eden CG, Lamme VA, Uylings HB. Heterotopic cortical afferents to the medial prefrontal cortex in the rat. A Combined Retrograde and Anterograde Tracer Study. European Journal of Neuroscience 1992;4:77–97. [16] Hoover WB, Vertes RP. Anatomical analysis of afferent projection to the medial prefrontal cortex in the rat. Brain Structure and Function 2007;212: 149–79. ´ D, Sobolewska A, Maciejak P, [17] Lehner M, Taracha E, Skórzewska A, Turzynska et al. Expression of c fos and CRF in the strength of fear response. Behavioural Brain Research 2008;188:154–67. ˛ scik [18] Lehner M, Wisłowska-Stanek A, Maciejak P, Szyndler J, Sobolewska A, Krza´ P, et al. Relationship between sensitivity and conditioned fear response in rats. Acta Neurobiologiae Experimentalis 2010;70:56–66. [19] Lehner M, Wisłowska-Stanek A, Taracha E, Maciejak P, Szyndler J, Skórzewska A, et al. The effects of midazolam and d-cycloserine on the release of glutamate and GABA in the basolateral amygdala of low and high anxiety rats during extinction trial of a conditioned fear test. Neurobiology of Learning and Memory 2010;94:468–80. [20] Barnes Jr EM. Intracellular trafficking of GABA(A) receptors. Life Sciences 2000;66:1063–70.
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35 [21] Nishi M, Kawata M. Dynamic of glucocorticoid receptor and mineralocorticoid receptor: implication from live cell imaging studies. Neuroendocrinology 2007;85:186–92. [22] Usuku T, Nishi M, Morimoto M, Brewer JA, Muglia LJ, Sugimoto T, et al. Visualization of glucocorticoid receptor in the brain of green fluorescent protein glucocorticoid receptor knockin mice. Neuroscience 2005;135:1119–28.
35
[23] Melcangi RC, Panzica G, Garcia-Segura LM. Neuroactive steroids: focus on human brain. Neuroscience 2011;191:1–5. [24] Van de Kar LD, Blair ML. Forebrain pathways mediating stressinduced hormone secretion. Frontiers in Neuroendocrinology 1999;20: 1–48.
Contents lists available at SciVerse ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communication
Corticosterone attenuates conditioned fear responses and potentiates the expression of GABA-A receptor alpha-2 subunits in the brain structures of rats selected for high anxiety b ´ A. Wisłowska-Stanek a,∗ , M. Lehner b , A. Skórzewska b , P. Maciejak a,b , J. Szyndler a , D. Turzynska , b a,b ´ A. Sobolewska , A. Płaznik a b
Department of Experimental and Clinical Pharmacology, Medical University of Warsaw, 26/28 Krakowskie Przedmie´scie Street, 00-927 Warsaw, Poland Department of Neurochemistry, Institute of Psychiatry and Neurology, 9 Sobieskiego Street, 02-957 Warsaw, Poland
h i g h l i g h t s Corticosterone decreased expression of conditioned fear in high anxiety rats (HR). Fear increased expression of alpha-2 subunits of GABA-A receptor in limbic structures. Corticosterone potentiated effects of conditioned fear on alpha-2 subunit expression.
a r t i c l e
i n f o
Article history: Received 30 May 2012 Received in revised form 6 July 2012 Accepted 10 July 2012 Available online 20 July 2012 Keywords: GABA-A receptor alpha-2 subunits Immunocytochemistry Brain structures Corticosterone Conditioned fear Individual differences
a b s t r a c t The aim of the experiment was to assess the effects of an acutely administered corticosterone on the expression of GABA-A receptor alpha-2 subunits in the brain structures of high (HR) and low (LR) anxiety rats (divided according to their conditioned fear-induced freezing response) subjected to a second conditioned fear session (1 week after fear conditioning). We found that corticosterone (20 mg/kg, sc) given to rats prior to the second conditioned fear session significantly enhanced a decrease in fear expression in the HR group. The behavioural effect of fear was accompanied by the increased expression of alpha-2 subunits in the basolateral amygdala (BLA) and the dentate gyrus of the hippocampus (DG) of the HR group. Corticosterone potentiated the effect of fear on alpha-2 subunit expression in the BLA, DG, the cingulate cortex area 1 and the secondary motor cortex (areas Cg1 and M2). The current study provides insight into the mechanisms that may be responsible for the beneficial effects of glucocorticoids in the therapy of some anxiety disorders. © 2012 Elsevier B.V. All rights reserved.
Acute glucocorticosteroid administration facilitates active behavioural coping in threatening situations [1]. We previously found that a single dose of corticosterone given to rats before training in a conditioned fear test significantly attenuated the freezing response when examined 24 h later [2,3]. Similarly, the acute administration of corticosterone before a session in an elevated plus maze produced an anxiolytic effect in rats [4]. It is also noteworthy that cortisol (20 mg) administered orally 1 h before each extinction-based session of psychotherapy significantly reduced anxiety symptoms during exposure to a phobic situation [5]. Recently, using a model of individual differences in fear responses of rats selected according to low and high freezing responses in the contextual fear test (defined as ‘low- and high
∗ Corresponding author. Tel.: +48 22 8262116; fax: +48 22 8262116. E-mail address: [email protected] (A. Wisłowska-Stanek). 0166-4328/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2012.07.018
anxiety’ rats; LR and HR groups, respectively), we found that HR rats had deficits in the activity of the brain structures that control the cognition necessary to cope with stress (i.e., the prefrontal cortex, as measured by c-Fos expression) and increased activity of the amygdalar nuclei that enhance the stress response (c-Fos/glucocorticoid receptors-ir) [6]. We also observed that some behavioural or pharmacological interventions attenuated the increased fear responses of HR rats. For example, we found that the administration of dcycloserine and midazolam before the testing session attenuated the freezing of HR rats and increased the expression of alpha-2 subunits of the GABA-A receptor in limbic areas and the prefrontal cortex [7]. Moreover, other preclinical data suggest that a decrease in conditioned fear is correlated with the upregulation of GABAergic markers (i.e., alpha-2 subunits of the GABA-A receptor) in the amygdala [8]. The role of alpha-2 subunits in fear processing is evidenced by the finding that mice with point-mutated alpha-2 GABA-A receptor subunits are resistant to the anxiolytic-like effects
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
of benzodiazepines and display a greater behavioural response to fear-conditioned stimuli [9]. In light of these data, we decided to test a hypothesis that the corticosterone-induced decrease in rat freezing responses is related to the drug and stress-induced changes in the expression of alpha-2 subunits in limbic and cortical areas. The experiments were performed on a cohort of 71 male Wistar rats (180–200 g body weight) purchased from a licenced breeder and housed in standard laboratory conditions under a 12 h light/dark cycle (lights on at 7 a.m.). The experiments were performed in accordance with the European Communities Council Directive of November 24th, 1986 (86/609 EEC). The Local Committee for Animal Care and Use at Warsaw Medical University, Poland approved all experimental procedures using animal subjects. After 4 days of acclimatisation, 40 animals (S group) were subjected to a conditioned fear test as previously described by Wisłowska-Stanek et al. [7]. In short, the experiment was performed over 3 consecutive days. On the first day, the animals were individually placed in a training box for 2 min to adapt to the experimental conditions. The following day, after 5 min of habituation to the training box, animals underwent a fear conditioning procedure that consisted of 3 footshocks (0.7 mA, 1 s, repeated every 59 s), and the animals were removed from the testing box 3 min after the last footshock was delivered. Conditioned fear was tested on the third day (test day, T1) by re-exposing rats to the testing box and recording the freezing response over 10 min. Freezing behaviour was measured by photo beams (10 Hz detection rate) controlled by a fear conditioning PC-program. The animals were divided according to their context-induced freezing responses into LR, low-anxiety rats with total durations of freezing responses one SEM or more below the mean (i.e., <224.1 s, mean = 246 s and SEM = 21.9 s), and an HR group consisting of high-anxiety rats with total durations of freezing responses one SEM or more above the mean (i.e., >267.9 s mean = 246 s, SEM = 21.9 s). These animals were defined as the LR-T1 (n = 18) and HR-T1 (n = 16) group, respectively. The control group, C (n = 6), was not conditioned but only placed in the conditioning boxes (mean freezing duration = 92.7 s, SEM = 24.45). After testing, animals remained undisturbed in their home cages for 7 days and were then subjected to the aversive context in a second fear testing session (T2). Before the T2 session, the pre-selected groups, LR-T1 and HR-T1, were further divided into the following subgroups: LR-T2-v, low-anxiety rats given vehicle
31
(n = 7); LR-T2-cort, low-anxiety rats given corticosterone (n = 9); HR-T2-v, high-anxiety rats given vehicle (n = 8); and HR-T2-cort, high-anxiety rats given corticosterone (n = 8). The corticosterone (20 mg/kg; Sigma–Aldrich, Poland) was suspended in sesame oil and administered subcutaneously (sc) to the nape of the neck 30 min before the T2 session. Control animals were administrated sesame oil (Sigma–Aldrich, Poland). All rats were decapitated 2 h after drug or vehicle administration. A control study was performed on a second group of animals (n = 25). The rats [LR < 230.7 s (256.4 − 25.7 s); HR > 282.1 s (256.4 + 25.7 s)] were divided into four groups: LR-v, low-anxiety rats given vehicle (n = 6); LR-cort, low-anxiety rats given corticosterone (n = 5); HR-v, high-anxiety rats given vehicle (n = 6); and HR-cort, high-anxiety rats given corticosterone (n = 8). The design of the second part of the experiment was the same as the first part of the experiment with the exception that the second fear test session was not performed (T2). Thus, these animals were injected with vehicle or corticosterone 7 days after T1 and decapitated 2 h later. Immunocytochemical staining for the alpha-2 subunit of the GABA-A receptor was performed on slide-mounted frozen brain sections. Staining and counting techniques have been described previously by Wisłowska-Stanek et al. [7]. Cells were counted in the following subregions: the cingulate cortex (areas 1 and 2), the secondary motor cortex (AP: 1.20; Cg1, Cg2, M2), the basolateral amygdala and the dentate gyrus of the hippocampus (AP – 3.14; DG, BLA) [10] (Fig. 1). Three slices from each section of the brain were taken. Western blot analyses performed previously with the same antibody confirmed the specific binding of this antibody to the alpha-2 subunit. Examples of alpha-2 subunit GABA-A receptor immunopositive cells have been published by Lehner et al. [11]. The behavioural and biochemical data are presented as the means ± the SEM. The differences between the conditioned group (S) and the control group (C) were analysed by Student’s t-test. The differences in freezing response durations during the first and second conditioned fear test (T1 and T2) in the LR and HR groups were analysed by one-way repeated measures ANOVA followed by Tukey’s post hoc test. The behavioural data from the second test session were analysed by two-way ANOVA followed by Tukey’s post hoc test. The immunocytochemical data were analysed by three-way ANOVA followed by Tukey’s post hoc test. For correlation analyses, a Pearson coefficient was calculated. Statistical analyses
Fig. 1. Diagrams adapted from Paxinos and Watson [10] showing regions of the brain analysed for expression of alpha-2 subunits. BLA: basolateral amygdala; Cg1, Cg2: cingulate cortex area 1 and 2; CPu: caudate putamen; DG: dentate gyrus of the hippocampus; M2 area: prefrontal cortex, secondary motor cortex. Shaded areas indicate the analysed brain regions.
32
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
Fig. 2. Freezing behaviour during the first fear conditioning test (T1) and second fear session (T2). (A) Behaviour of saline pretreated rats. LR-T1: low-anxiety rats, first contextual fear test (n = 18); HR-T1: high-anxiety rats, first contextual fear test (n = 16); LR-T2-v: low anxiety rats, second contextual fear session (n = 7); HR-T2-v: high-anxiety rats, second contextual fear session (n = 8). (B) Freezing duration during the second fear session after pretreatment of rats with corticosterone or saline (T2). LR-T2-v: low-anxiety rats pretreated with vehicle (n = 7); HR-T2-v: high-anxiety rats, pretreated with vehicle (n = 8); LR-T2-cort: low-anxiety rats pretreated with corticosterone (n = 9); HR-T2-cort: high-anxiety rats pretreated with corticosterone (n = 8). Data are shown as the means ± SEM. *Differs from LR-T1, **p < 0.01; & differs from LR-T2-v, && p < 0.01; # differs from HR-T2-v, # p < 0.05.
were performed using Statistica 8.0 for Windows (StatSoft Inc., USA). A Student’s t-test revealed a significant difference in freezing response durations between the conditioned (S) and control (C) groups in the conditioned fear test; i.e., the effect of contextual fear conditioning; t = 2.66, df = 44, P = 0.01. ANOVA revealed significant differences in freezing durations between the first (T1) and second fear (T2) sessions and significant effects of group [F(1,15) = 65.39; p < 0.01] and time [F(1,15) = 7.79; P = 0.01], but there was no significant group × time interaction effect [F(1,15) = 0.68; P = 0.42]. Post hoc tests revealed longer freezing durations in the HR-T1 group compared to the LR-T1 group (P < 0.01) and in the HR-T2-v group compared to the LR-T2-v group (p < 0.01) (Fig. 2A). ANOVA showed significant differences in freezing durations between HR and LR rats after vehicle and corticosterone administration during the T2 session: a significant effect of group [F(1,30) = 11.59; P < 0.01], a significant group × drug interaction effect [F(1,30) = 5.14; P < 0.05] and no drug effect [F(1,30) = 3.80; P = 0.06]. Post hoc tests showed significantly longer freezing durations in the HR-T2-v group compared with the LR-T2-v group (P < 0.01) and a significant decrease
in freezing durations in the HR-T2-cort rats compared to HR-T2-v rats (P < 0.05) (Fig. 2B). ANOVA showed statistically significant differences between groups in the density of cells expressing alpha-2 GABA-A receptor subunits in Cg1, M2, BLA and DG (Table 1). Conditioned fear enhanced the expression of alpha-2 in the BLA and DG of the HRT2-v group (P < 0.01) and in the DG and Cg1 of the LR-T2-v rats (P < 0.01). The steady-state levels of alpha-2 subunits were the same in both groups; however, after the second fear session, the concentration was higher in the BLA (P < 0.01) and lower in the DG (P < 0.01) of HR-T2-v rats than LR-T2-v rats. Tukey’s post hoc test revealed higher expression of alpha-2 subunits in the HR-T2-cort group compared to the HR-T2-v and HR-cort groups in the Cg1, M2, BLA and DG (P < 0.01). Moreover, the LR-T2-cort group had higher expression of alpha-2 subunits in M2 (P < 0.01) and the DG (P < 0.05) compared to the LR-T2-v group. Post hoc analyses also revealed enhanced expression of alpha-2 subunits in Cg1, M2, DG (P < 0.01) and BLA (P < 0.05) in LR-T2-cort group compared to the LR-cort group. Correlation analyses revealed a significant negative relationship between freezing times during the T2 session and the expression of alpha-2 subunits in area M2 [r = (−)0.61; P = 0.01], the BLA [r = (−)0.70; P = 0.01] and the DG [r = (−)0.59; P = 0.02] in the HR group. In the LR group, correlation analyses did not reveal any significant effects. The main finding of this study shows that corticosterone given to rats prior to a second exposure to an aversive context significantly enhanced a spontaneously occurring decrease in fear expression in the HR group. The expression of alpha-2 subunits was increased by contextual fear in the BLA and DG of the HR group and in Cg1 and the DG of LR animals. Corticosterone potentiated the effect of fear on alpha-2 subunits in all examined structures except Cg2 in the HR group (HR-T2-cort vs. HR-T2-v). A less potent effect was present in the LR-T2-cort vs. LR-T2-v group (DG and M2 area). Importantly, pretreatment of rats with corticosterone alone did not change the expression of alpha-2 subunits in the control groups that were not exposed to an aversive context. These data implicate the role of GABAergic innervation of cortical and limbic structures in the central effects of corticosterone on fear responses. The amygdala, prefrontal cortex and hippocampus have long been recognised as a neural system responsible for the generation and elaboration of conditioned fear [cf. 12]. It is well known that this neural substrate of emotions is under tonic inhibitory control by GABAergic mechanisms [13]. The GABAergic system is not only important for the acquisition of conditioned fear but also in the attenuation of fear responses. In an experiment that evaluated training-induced changes in the expression of GABA-A-associated genes in the amygdala during the extinction of Pavlovian fear, fear extinction increased the levels of the alpha-2 and beta-2 GABA-A receptor subunit and GAD67 mRNA and decreased GABA transporter-1 mRNA [8]. These results show that fear attenuation coincides temporally with an upregulation of GABAergic markers related to enhanced GABAergic transmission. The prefrontal cortex may influence contextual fear conditioning by contributing to the integration of information about the encoded environment via the amygdala and contextual information from the hippocampus [14]. The role of the medial prefrontal cortex in this process has been documented. Van Eden et al. have shown with anterograde labelling that Fr2 (M2 area) and the anterior cingulate area (Acd) receive more projections from somatosensory and associational visual cortices than from the primary motor cortex [15]. Furthermore, the study by Hoover and Vertes [16] indicates that the information from widespread areas of the cortex is presumably integrated by the dorsal prefrontal cortex (the medial frontal agranular areas (AGm-Fr2-M2) and the anterior cingulate (ACd)) during goal directed action.
Table 1 Alpha-2 subunits of GABA-A receptor immunostaining after drug injection and the re-exposition to contextual fear test.
Cg1
HR-v (n = 6)
HR-cort (n = 8)
LR-v (n = 6)
LR-cort (n = 5)
HR-T2-v (n = 8)
HR-T2-cort (n = 8)
222.1 ± 7.6
192.1 ± 13.9
203.3 ± 6.5
203.4 ± 14.6
262.3 ± 24.2
350.8 ± 20.1bb, **
LR-T2-v (n = 7) 327.4 ± 16.6aa
LR-T2-cort (n = 9)
ANOVA
320,5 ± 18.3bb
1
[F(1,44) = 0.33 (P = 0.56)] [F(1,44) = 1.18 (P = 0.28)] 3 [F(1,44) = 86,24 (P < 0.01)] 4 [F(1,44) = 1.9 (P = 0.17)] 5 [F(1,44) = 0.79 (P = 0.37)] 6 [F(1,44) = 5,54 (P < 0.05)] 7 [F(1,44) = 7.01 (P = 0.01)] 2
235.0 ± 9.4
234.7 ± 22.2
216.6 ± 7.6
198.8 ± 10.9
216.6 ± 7.6
228.7 ± 12.5
236.3 ± 14.5
230.7 ± 12.0
1
[F(1,50) = 1.3 (P = 0.25)] [F(1,50) = 0.39 (P = 0.53)] [F(1,50) = 1.03 (P = 0.31)] 4 [F(1,50) = 0.29 (P = 0.59)] 5 [F(1,50) = 2.42 (P = 0.12)] 6 [F(1,50) = 0.07 (P = 0.78)] 7 [F(1,50) = 0.11 (P = 0.73)] 2 3
M2
145.0 ± 6.7
149.4 ± 5.8
135.0 ± 7.4
134.3 ± 6.5
181.6 ± 15.7
355.1 ± 16.8bb, **,##
143.3 ± 16.8
270.0 ± 14.9bb,&&
1
[F(1,49) = 15.09 (P < 0.01)] [F(1,49) = 63.75 (P < 0.01)] 3 [F(1,49) = 102.11 (P < 0.01)] 4 [F(1,49) = 1.74 (P = 0.19)] 5 [F(1,49) = 6.63 (P < 0.05)] 6 [F(1,49) = 59.611 (P < 0.01)] 7 [F(1,49) = 1.27 (P = 0.26)] 2
BLA
153.0 ± 14.1
168.9 ± 12.2
155.2 ± 14.1
162.8 ± 15.4
260.6 ± 12.8aa,&&
333.1 ± 11.9bb, **,##
194.00 ± 12.0
212.1 ± 11.3b
1
[F(1,49) = 26.98 (P < 0.01)] [F(1,49) = 9.59 (P < 0.01)] 3 [F(1,49) = 95.34 (P < 0.01)] 4 [F(1,49) = 2.89 (P = 0.09)] 5 [F(1,49) = 24.82 (P < 0.01)] 6 [F(1,49) = 3.32 (P = 0.07)] 7 [F(1,49) = 1.57 (P = 0.22)] 2
DG
104.0 ± 4.5
100.2 ± 3.4
91.8 ± 4.8
94.8 ± 4.6
155.9 ± 12.8aa,&&
252.1 ± 13.7bb, **
204.7 ± 12.8aa
237.7 ± 14.8bb,&
1
[F(1,49) = 0.28 (P = 0.59)] [F(1,49) = 16.76 (P < 0.01)] 3 [F(1,49) = 214.57 (P < 0.01)] 4 [F(1,49) = 3.24 (P = 0.07)] 5 [F(1,49) = 2.75 (P = 0.10)] 6 [F(1,49) = 17.17 (P < 0.01)] 7 [F(1,49) = 4.99 (P < 0.05)]
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
Cg2
2
Data are shown as means ± SEM (cell number/1 mm2 ). 1 Statistically significant group effect, 2 statistically significant drug effect, 3 statistically significant fear effect, 4 statistically significant group × drug interaction, 5 statistically significant group × fear interaction, 6 statistically significant drug × fear interaction, 7 statistically significant group × drug × fear interaction. a Differs from HR-v or LR-v, respectively, aa P < 0.01; b differs from HR-cort or LR-cort, respectively, bb P < 0.01; *differs from HR-T2-v, **P < 0.01; & differs from LR-T2-v, & P < 0.05; && P < 0.01; # differs from LR-T-cort, ## P < 0.01. For more details see section ‘Materials and methods’.
33
34
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35
These findings were extended by our previous results indicating that LR animals had greater activity (as measured by c-Fos expression) of the dorsomedial part of the prefrontal cortex (area M2) than the HR group [6]. It is also noteworthy that HR and LR rats have different fear coping mechanisms. Contextual fear exposition caused less freezing and more aversive ultrasonic vocalisation in the LR group compared to HR rats [17]. The less anxious behaviour of the LR rats was not related to a higher pain threshold [18]. Moreover, our data excluded a role for deficits in cognitive processes because no differences between LR and HR rats were found in the Morris water maze (unpublished data). Recently, we also showed that the less anxious behaviour of LR animals given saline was accompanied by elevated basal levels of glutamate in the BLA and a stronger elevation of GABA in response to contextual fear when compared to HR rats. Pretreatment of rats with d-cycloserine and midazolam, drugs used in the treatment of post-traumatic stress disorder, significantly increased the concentration of GABA in the BLA and inhibited the expression of contextual fear in the HR group [19]. Moreover, these drugs caused an increase in alpha-2 subunits in limbic and cortical areas [7]. In the current study, a significant decrease in the freezing responses in the second fear session after corticosterone administration in the HR animals was accompanied by an increase in the expression of alpha-2 subunits in Cg1 and M2 of the prefrontal cortex and the DG and BLA. Thus, a drug- or corticosterone-induced increase in the expression of alpha-2 subunits in limbic structures and prefrontal cortical areas in HR rats may be a compensatory mechanism that facilitates active coping in a threatening situation. The increase in the expression of the alpha-2 subunits could be the result of enhanced receptor protein synthesis or stimulation of their active transport to the cell surface, a process that may be regulated by protein kinase C [20]. Corticosterone can significantly contribute to the rapid cycling of GABA-A receptors between the cell surface and the subsynaptic pool, which would provide a mechanism for the short-term regulation of GABAergic neurotransmission [20,21]. It has been demonstrated that glucocorticoid and mineralocorticoid receptors (GRs and MRs) are quickly translocated from the cytoplasm to the nucleus after corticosterone treatment through association with importins [21]. In the nucleus, the interaction of GRs with glucocorticoid response elements and transcriptional components and receptor clearance are dynamically regulated within seconds and minutes [22]. These studies indicate that glucocorticoids may rapidly modify the process of nuclear activation and the production of various receptor proteins and their intracellular cycling. However, the exact molecular nature of the changes in the alpha-2 subunit expression after corticosterone remains unclear, and the literature lacks a direct reference to this problem. Another rapid, non-genomic, mechanism of corticosterone action might involve the formation of GABA-A receptor-modulating neurosteroids from a corticosterone derivative, deoxycorticosterone. These derivatives, including allopregnanolone, can bind to GABA-A receptors composed of alpha-2 subunits and evoke benzodiazepine-like effects, e.g., anxiolytic-like actions [23]. GABAA receptor modulators may alter the neuroendocrine stress response, which can affect cortical control over the limbic nuclei and can also affect emotional memory [24]. In conclusion, the current study shows that animals that are more vulnerable to fear-evoking stimuli differ in the mechanisms that control GABA-A receptor alpha-2 subunit expression in prefrontal cortical areas and limbic structures. Further, our results implicate glucocorticoid-related mechanisms in this phenomenon. The current study also provides insight into the central mechanisms responsible for the beneficial effects of glucocorticoids in the therapy of some anxiety disorders. It should be also noted that similar effects have previously been found after midazolam,
d-cycloserine and corticosterone administration, indicating that their mechanisms of action in anxiety disorders are likely similar.
Acknowledgements This study was supported by statutory Grant No. 50100312043 from the Institute of Psychiatry and Neurology and by Grant No. 0440/B/P01/2009/36 from the Ministry of Science and Higher Education (ML), Warsaw, Poland.
References [1] Clay R, Hebert M, Gill G, Stapleton LA, Pridham A, Coady M, et al. Glucocorticoids are required for extinction of predator stress-induced hyperarousal. Neurobiology of Learning and Memory 2011;96:367–77. ´ ´ [2] Skórzewska A, Bidzinski A, Lehner M, Turzynska D, Wisłowska-Stanek A, Sobolewska A, et al. The effects of acute and chronic administration of corticosterone on rat behavior in two models of fear responses, plasma corticosterone concentration, and c-Fos expression in brain structures. Pharmacology Biochemistry and Behavior 2006;85:522–34. ´ ´ A, Lehner M, Turzynska D, Sobolewska A, Hamed A, [3] Skórzewska A, Bidzinski et al. The effects of acute corticosterone administration on anxiety, endogenous corticosterone, and c-Fos expression in the rat brain. Hormones and Behavior 2007;52:317–25. [4] Andreatini R, Leite JR. The effect of corticosterone in rats submitted to the elevated plus-maze and to pentylenetetrazol-induced convulsions. Progress in Neuro-Psychopharmacology and Biological Psychiatry 1994;18: 1333–47. [5] de Quervain DJ, Bentz D, Michael T, Bolt OC, Wiederhold BK, Margraf J, et al. Glucocorticoids enhance extinction-based psychotherapy. Proceedings of the National Academy of Sciences of the United States of America 2011;108:6621–5. ´ [6] Lehner M, Taracha E, Maciejak P, Szyndler J, Skórzewska A, Turzynska D, et al. Colocalisation of c-Fos and glucocorticoid receptor as well as 5-HT1A and glucocorticoid receptor immunoreactivity-expressing cells in the brain structures of low and high anxiety rats. Behavioural Brain Research 2009;200: 150–9. [7] Wisłowska-Stanek A, Lehner M, Skórzewska A, Maciejak P, Szyndler J, ´ Turzynska D, et al. Effects of d-cycloserine and midazolam on the expression of the GABA-A alpha-2 receptor subunits in brain structures of fear conditioned rats. Behavioural Brain Research 2011;225:655–9. [8] Heldt SA, Ressler KJ. Training-induced changes in the expression of GABA-A associated genes in the amygdala after acquisition and extinction of Pavlovian fear. European Journal of Neuroscience 2007;26:3631–44. [9] Morris HV, Dawson GR, Reynolds DS, Atack JR, Stephens DN. Both alpha2 and alpha3 GABAA receptor subtypes mediate the anxiolytic properties of benzodiazepeine site ligands in the conditioned emotional paradigm. European Journal of Neuroscience 2006;23:2495–504. [10] Paxinos G, Watson CH. The rat brain in stereotaxic coordinates. San Diego, CA: Academic Press Inc; 1998. [11] Lehner M, Wisłowska-Stanek A, Skórzewska A, Maciejak P, Szyndler J, ´ D, et al. Differences in the density of GABA-A receptor alpha-2 subTurzynska units and gephyrin in brain structures of rats selected for low and high anxiety in basal and fear-stimulated conditions, in a model of contextual fear conditioning. Neurobiology of Learning and Memory 2010;94:499–508. [12] Gray J. A theory of anxiety: the role of the limbic system. Encephale 1983;9:161B–6B. [13] Santos JM, Macedo CE, Brandao ML. Gabaergic mechanisms of hypothalamic nuclei in the expression of conditioned fear. Neurobiology of Learning and Memory 2008;90:560–8. [14] Hurley KM, Herbert H, Moga˛ MM, Saper CM. Efferent projection of infralimbic cortex of the rat. Journal of Comparative Neurology 1991;308:249–76. [15] Van Eden CG, Lamme VA, Uylings HB. Heterotopic cortical afferents to the medial prefrontal cortex in the rat. A Combined Retrograde and Anterograde Tracer Study. European Journal of Neuroscience 1992;4:77–97. [16] Hoover WB, Vertes RP. Anatomical analysis of afferent projection to the medial prefrontal cortex in the rat. Brain Structure and Function 2007;212: 149–79. ´ D, Sobolewska A, Maciejak P, [17] Lehner M, Taracha E, Skórzewska A, Turzynska et al. Expression of c fos and CRF in the strength of fear response. Behavioural Brain Research 2008;188:154–67. ˛ scik [18] Lehner M, Wisłowska-Stanek A, Maciejak P, Szyndler J, Sobolewska A, Krza´ P, et al. Relationship between sensitivity and conditioned fear response in rats. Acta Neurobiologiae Experimentalis 2010;70:56–66. [19] Lehner M, Wisłowska-Stanek A, Taracha E, Maciejak P, Szyndler J, Skórzewska A, et al. The effects of midazolam and d-cycloserine on the release of glutamate and GABA in the basolateral amygdala of low and high anxiety rats during extinction trial of a conditioned fear test. Neurobiology of Learning and Memory 2010;94:468–80. [20] Barnes Jr EM. Intracellular trafficking of GABA(A) receptors. Life Sciences 2000;66:1063–70.
A. Wisłowska-Stanek et al. / Behavioural Brain Research 235 (2012) 30–35 [21] Nishi M, Kawata M. Dynamic of glucocorticoid receptor and mineralocorticoid receptor: implication from live cell imaging studies. Neuroendocrinology 2007;85:186–92. [22] Usuku T, Nishi M, Morimoto M, Brewer JA, Muglia LJ, Sugimoto T, et al. Visualization of glucocorticoid receptor in the brain of green fluorescent protein glucocorticoid receptor knockin mice. Neuroscience 2005;135:1119–28.
35
[23] Melcangi RC, Panzica G, Garcia-Segura LM. Neuroactive steroids: focus on human brain. Neuroscience 2011;191:1–5. [24] Van de Kar LD, Blair ML. Forebrain pathways mediating stressinduced hormone secretion. Frontiers in Neuroendocrinology 1999;20: 1–48.