Mesolimbic dopamine D2 receptor plasticity contributes to stress resilience in rats subjected to chronic mild stress

Few studies have investigated neurobiological and biochemical differences between stress-resilient and stress-vulnerable experimental animals. We investigated alterations in mesolimbic dopamine D2 receptor density and mRNA expression level in stressed rats at two time points, i.e. after 2 and 5 weeks of chronic mild stress (CMS). We used the chronic mild stress paradigm because it is a well-established animal model of depression. Two groups of stressed rats were distinguished during CMS experiments: (1) stress reactive (70 %), which displayed a decrease in the drinking of a palatable sucrose solution during the stress regimen, and (2) stress resilient (30 %), which exhibited an unaltered drinking profile when compared with the unchallenged control group. [3H]Domperidone was used as a ligand to label dopamine D2 receptors, and a mixture of three specific oligonucleotides was used to evaluate dopamine D2 receptor mRNA changes in various regions of the rat brain. CMS strongly affected the mesolimbic dopamine circuit in stress-resilient group after 2 weeks and stress-reactive group of rats after 5 weeks which exhibited a decrease in the level of dopamine D2 receptor protein without alterations in D2 mRNA expression. Stress-resilient animals, but not stress-reactive animals, effectively adapted to the extended stress and coped with it. The increase in D2 mRNA expression returned the dopamine D2 receptor density to control levels in stress-resilient rats after 5 weeks of CMS, but not in stress-reactive animals. These results clearly demonstrate that, despite earlier blunting, the activation of dopamine receptor biosynthesis in the dopamine mesoaccumbens system in stress-resilient rats is involved in active coping with stressful experiences, and it exhibits a delay in time.


Introduction
Psychological resilience among people in the face of trauma, adversity and chronic stress has received much attention since the 1970s (Feder et al. 2009). Many behavioural and psychological aspects have been investigated, and differences have been observed between stress-resilient and stress-vulnerable people. The stress-resilient group exhibits active coping in the face of trauma and an optimistic mood, and they cognitively reinterpret stressful situations and may possess a strong moral compass and life purpose (Haglund et al. 2007;Feder et al. 2009;Cicchetti 2010;Rowland 2011).
Advanced biological, molecular and brain imaging techniques as well as animal models of mental illnesses have allowed the delineation of multiple genetic, biochemical, neurological, neurochemical and neuroendocrine changes that define the stress-vulnerable phenotype. However, relatively few studies have examined differences in neurobiological systems among individuals who are "resistant" to stress. Therefore, further intensive study is required to define the molecular factors and aspects of resilience.
Noradrenergic and serotonergic catecholamine circuits are imbalanced in stress-evoked depression. This phenomenon is well established, and these circuits are the main target of action of antidepressant drugs. These drugs immediately influence catecholamine concentrations, but their clinical and behavioural effects appear after several weeks of chronic administration. These two circuits are indirectly involved in mood control via ascending innervations that regulate the mesolimbic dopamine system (Willner et al. 2005;Dunlop and Nemeroff 2007). However, the dopaminergic system has received less attention regarding vulnerability and resilience to depression under different circumstances. The mesolimbic dopamine circuit, particularly dopaminergic projections from the ventral tegmental area (VTA) to the nucleus accumbens septi (NAcc), is the downstream terminal link in the nerve cascade that regulates reward-and aversionrelated behaviours (Gershon et al. 2007;Krishnan et al. 2007). Stress exerts a strong influence on this system and elicits specific responses depending on the type and duration of aversive stimulation (Puglisi-Allegra et al. 1991;Puglisi-Allegra 1994, 1996a, b;Finlay and Zigmond 1997). Therefore, impairment of the stress-induced mesolimbic dopamine system leads to the development of anhedonia, which is a core symptom of depression. The rapid antidepressant action of dopamine D 2 receptor agonists, such as pramipexole or bromocriptine, likely explains the direct involvement of the mesolimbic circuit and dopamine D 2 receptors in depression (Dailly et al. 2004). Alterations in dopamine concentration in the mesolimbic reward system and dopamine D 2 receptors in response to stress accompany the development of depressionlike psychopathology (Imperato et al. 1993;Papp et al. 1994;Dziedzicka-Wasylewska et al. 1997;Larisch et al. 1997;Cabib et al. 1998;Yadid et al. 2001;Gershon et al. 2007). Dopamine concentrations in the NAcc are increased in response to and in expectation of reward or aversive stimuli, but a blunted response leads to anhedonia, which is a long-lasting effect of mild chronic stress (Di Tanda 1997, Di Chiara et al. 1999;Dunlop and Nemeroff 2007). Our study used the chronic mild stress (CMS) paradigm, which is a well-characterised animal model of depression with good face and predictive and construct validity. The chronic exposure of rats to mild unpredictable stressors produces behavioural deficits (anhedonia) that may be maintained for several months, and chronic treatment with antidepressants restores normal behaviour in stressed animals (Willner et al. 1987;Papp et al. 1994;Bielajew et al. 2002;Bekris et al. 2005). Approximately 30 % of animals that are subjected to several weeks of CMS demonstrate stress resilience that is manifested as unaltered consumption of a palatable 1 % sucrose solution in this model (Bergstrom et al. Bergström et al. 2008). These animals are usually rejected from further analysis as "bad-responding" subjects that do not demonstrate the behavioural response. However, this group mimics the behavioural response to stress that is observed in people who do not develop depression in the face of adversity. Therefore, the present study investigated differences in the dopaminergic mesolimbic system, especially dopamine D 2 receptor density and mRNA expression, in various brain regions between stressreactive and stress-resilient rats which followed CMS paradigm. The novel approach in our study involves the follow-up of temporal changes in the dopamine circuit during the CMS paradigm in both groups of animals, after 2 and 5 weeks of the stress procedure. This study clearly demonstrates that resilient animals exhibit strong neuroadaptive abilities that allow them to dynamically and actively cope with stress and adversity.

Animals
Male Wistar Han rats (Charles River, Germany) were brought into the laboratory 1 month prior to the start of the behavioural and biochemical experiments and, with some exceptions described below, were singly housed in plastic cages (40×25× 15 cm) with food and water provided ad libitum on a 12h light/dark cycle under constant temperature (22±2°C) and humidity (50±5 %). The animals weighed between 300 and 400 g at the end of the behavioural experiments.
Behavioural procedure and tissue preparation Two CMS experiments were performed according to the method described by Papp et al. (1994). All animals (n= 100) were adapted to laboratory and housing conditions for 2 weeks and trained to a drink palatable sucrose solution (1 %). The training procedure lasted 6 weeks and consisted of 1-h testing sessions every week (at 10:00 AM on Tuesdays) in which the sucrose solution was presented to the rats in their home cages after 14 h of food and water deprivation. Sucrose intake was measured after each drinking test as differences in bottle weight. Ten animals were excluded from the experiment after the 6-week training procedure because of non-stable drinking profiles. The remaining animals (n=90) were randomly divided into two groups and subjected to control (n=20) or stress (n=70) conditions. The stress regimen was initiated at week 6. Half of the stressed animals (n=35) were subjected to the chronic mild stress procedure for 2 weeks, and the other half (n=35) of the stressed animals underwent CMS for five consecutive weeks. The stress schedule included two periods of food or water deprivation, a 45°cage tilt, intermittent illumination, a soiled cage (200 ml of water in the sawdust bedding), paired housing, low-intensity stroboscopic illumination (150 flashes/min) and two periods of no stress. All stressors had a duration of 10-14 h and were applied individually and continuously during light and dark periods. Control animals were remained undisturbed in a separate room with free access to food and water, except for a period of overnight deprivation for the sucrose consumption test once per week. Three different groups of animals were identified based on sucrose intake after 2 and 5 weeks of the CMS procedure: -2-week CMS: (1) control animals (n=10), (2) stressreactive rats (n=25) and (3) stress-resilient rats (n=10); -5-week CMS: (1) control animals (n=10), (2) stressreactive rats (n=24) and (3) stress-resilient rats (n=11).
Thirty percent of all rats that were subjected to the chronic mild stress procedure were stress resilient (n=21). Ten representative stress-reactive animals from the 2-and 5-week CMS were selected for biochemical analyses. The remaining stressreactive animals (n=29) were used in another experiment, and these animals were not included in this study.
The rats were sacrificed by decapitation 24 h after the last sucrose test. The brains were rapidly removed and frozen using a heptane-dry ice mixture. Coronal brain sections (12 μm) through the nucleus accumbens septi, striatum and ventral tegmental area (Paxinos and Watson 1998) were cut using a Jung CM 3000 cryostat microtome (Leica, Germany). The slices were thaw mounted on gelatine-covered microscope slides, air dried and stored at −20°C until use.
[ 3 H]Domperidone binding to dopamine D 2 receptors and analysis of autoradiograms Tissue sections were pre-incubated in 50 mM Tris-HCl buffer (pH7.4) at room temperature for 15 min to remove endogenous dopamine. The brain slices were incubated for 2 h at room temperature in 50 mM Tris-HCl (pH 7.4) incubation buffer containing 0.4 nM [ 3 H]domperidone and the following ions: 120 mM NaCl, 1 mM EDTA, 1.5 mM CaCl 2 , 4 mM MgCl 2 and 5 mM KCl. [ 3 H]Domperidone concentrations corresponded to Kd value (Seeman et al. 2003). Parallel sections treated with 10 μM (+)butaclamol were incubated in the buffer described above to determine non-specific binding. Incubation was terminated by washing the sections twice in 50 mM Tris-HCl (pH7.4) at 4°C for 5 min and once in ice-cold distilled water. The sections were dried overnight under a stream of air. The labelled brain slices were placed against an imaging plate (Fujifilm, Japan) with autoradiographic microscales (GE Healthcare) for 7 days. The obtained autoradiograms were analysed and quantified using ImageGauge software (Fujifilm, Japan). The specific binding of radioligand to dopamine D 2 receptors was calculated by subtracting non-specific binding images in adjacent brain slices from the total binding signal. The results are expressed in femtomole of bound radioligand per milligram of tissue (fmol/mg) in each examined structure.

Dopamine D 2 receptor mRNA in situ hybridisation
The rat brain sections were fixed for 10 min in cold 4 % formaldehyde, briefly washed in PBS and incubated for 10 min in an ice-cold acetic anhydride (0.25 %)-TEA (0.1 M) solution-to minimise unspecific oligonucleotide hybridisation. The fixed brain slices were dehydrated in a graded series of alcohols and subjected to two 10-min incubations in chloroform to remove lipids. The prepared tissue slices were washed in ethanol and air dried. A mixture of three commercially available oligonucleotides complementary to 4-51, 766-813 and 848-901 bp of the rat dopamine D 2 receptor mRNA was used in the in situ hybridisation assays (Dziedzicka-Wasylewska et al. 1997;Bunzow et al. 1988). Oligonucleotide probes were labelled at the 3′ end with [ 35 S]dADP (Hartmann Analytic, Germany) using terminal transferase (Fermentas, Lithuania). The probes were suspended to a final concentration of 1×10 6 disintegration per minute (dpm) per 50 μl of hybridisation buffer containing the following components: 50 % (v/v) formamide, 10 % (w/v) dextran sulphate, 4× saline-sodium citrate (SSC) (pH7.0), 1× Denhardt's solution (0.02 % polyvinylpyrrolidone/0.02 % Ficoll/0.02 % bovine serum albumin), yeast tRNA (0.25 mg/ml), sheared herring sperm DNA (0.2 mg/ml) and 10 mM dithiotreitol. All solutions were prepared in deionised 0.1 % DEPCtreated water. Hybridisation buffer containing 50 μl of 1×10 6 dpm of radiolabelled oligonucleotides was applied to each fixed and prehybridised slide, and the slides were covered by Parafilm and incubated for 18 h at 37°C in humid conditions. Each slide was briefly washed twice in a 1× SSC solution at room temperature after hybridisation followed by four washes (15 min each) in 2× SSC with 50 % formamide at 42°C and one for 15-min wash at room temperature in 1× SSC. The hybridised brain sections were rinsed with deionised water, dehydrated in increasing concentrations of ethanol and air dried. Prepared tissue sections were placed into X-ray cassettes and exposed to film plates (Kodak) for 20 days at a temperature of −20°C. The developed autoradiograms were analysed, quantified and normalised using ImageGauge software (Fujifilm, Japan). Data are expressed as percent changes in mRNA signals compared with the control group (expressed as 100 %).

Statistical analysis of the data
The values from the CMS behavioural tests represent the mean sucrose intake at the corresponding time point. Twoway ANOVA and independent repeated measures ANOVA followed by Bonferroni's post hoc test were used to analyse the data from the 6 weeks of sucrose training and the 2 and 5 weeks of CMS. [ 3 H]Domperidone-binding data are expressed as the mean density of dopamine D 2 receptor protein per milligram of tissue [in femtomole per milligram ± standard error of the mean (SEM)] in each group. Data from in situ hybridisation assays were normalised to the data of the control group (expressed as 100 %). These data were analysed using one-way ANOVA and Bonferroni's multiple comparison post hoc test by using GraphPad Prism software (GraphPad Software, Inc., La Jolla, CA, USA).

Sucrose intake
Independent analysis using repeated measures ANOVA test did not show any significance of sucrose consumption among animals during training procedure in 2-and 5-week CMS experiment (F 2, 29 =1.190; p=0.3271, and F 2, 29 = 3.485; p =0.0526, respectively). After the CMS experiments, we were able to ascribe the animals retrospectively to the stress-resilient and stress-susceptible groups on the basis of their individual drinking behaviour. Two-way ANOVA revealed significant effects of stress (F 14, 216 = 17.51; p < 0.001), time (F 14, 216 = 2.630; p < 0.05) and stress×time (F 14, 216 =2.531; p< 0.005) on sucrose consumption after 2 weeks of CMS. A significant effect of stress (F 20, 297 =44.46; p<0.0001) and stress×time (F 20, 297 =2.303; p<0.005) on sucrose consumption was observed after 5 weeks of CMS in the respective groups of rats. Repeated measures ANOVA revealed that the application of the stress procedure for 2 weeks allowed the selection of stress-reactive animals (F 2, 29 = 14.74; p < 0.005) which exhibited a significant decrease in sucrose consumption as compared to non-stressed control group. A separate group of rats did not change their drinking behaviour when compared with control rats (F 2, 29 =14.74; p>0.05) (Fig. 1a).
A similar response was observed in the 5-week CMS rats. A proportion of the animals responded to stress with a decrease in sucrose consumption (F 2, 29 =10.79; p<0.05), and other animals were unaffected by stress (F 2, 29 =10.79; p>0.05) as compared to control rats (Fig. 1b). Our arbitrary retrospective observations have shown that anhedonic animals decreased their sucrose consumption to below 7.5 g when compared with the sucrose intake measurement performed at the end of the sucrose training period (i.e. sixth week).

WEEKS
Effect of CMS procedure on dopamine D 2 receptors in the rat brain-autoradiography and in situ hybridisation Representative autoradiograms of the distribution of [ 3 H] domperidone-binding sites and dopamine D 2 receptor mRNA in various regions of the rat brain are presented in Fig. 2. The impact of the CMS procedure on dopamine D 2 receptors was examined in several rat brain areas. Two weeks of CMS reduced [ 3 H]domperidone binding in the striatum (medial and lateral), nucleus accumbens (shell and core) and the lateral, but not medial, ventral tegmental area. However, this reduction only reached statistical significance in animals that were resilient to CMS procedure ( Fig. 3a-f). Two weeks of CMS did not alter dopamine D 2 receptor mRNA expression in the examined brain regions (Fig. 4). Five weeks of the CMS procedure reduced [ 3 H]domperidone binding to dopamine D 2 receptors in the striatum (lateral and medial) and nucleus accumbens (core) in stress-responsive rats. No significant differences in [ 3 H] domperidone binding were observed in the brains of stress-resilient animals (Fig. 5a, b, d). However, the reduction in [ 3 H]domperidone binding in the nucleus accumbens shell was observed in both stress-reactive and stress-resilient rats (Fig. 5c) No significant differences in [ 3 H]domperidone binding were observed in the ventral tegmental area when compared with the control group. However, a statistically significant difference was observed in dopamine D 2 receptor density between stress-reactive and stress-resilient animals in the medial part of the VTA (Fig. 5e, f).
In situ hybridisation revealed a significant increase in dopamine D 2 receptor mRNA in the striatum (lateral and medial), nucleus accumbens (shell) and ventral tegmental area (lateral and medial) in stress-resilient rats (Fig. 6a-f). No significant differences were observed in stress-reactive animals.

Discussion
Recent research has focused on the mechanisms that underlie the susceptibility to stress (Larish et al. Larisch et al. 1997;Husseini et al. 2001;Zhu et al. 2011;Blugeot et al. 2011). However, only a small percentage of people who face adverse events develop depression (Feder et al. 2009;Cicchetti 2010). Therefore, the factors that contribute to stress resilience may be more worthy of investigation. The present study examined stress-induced changes in mesolimbic dopamine D 2 receptor protein and mRNA using chronic mild stress at two time points (after 2 and 5 weeks of CMS) in two groups of animals with different behavioural responses to stress.
The CMS paradigm is a well-established animal behavioural model of anhedonia, which is a core symptom of depression. Male rodents are used in this model because males are more biologically stable than females. The dopamine reward circuit in female rats is strongly affected by hormonal changes during the oestrus cycle (Becker 1999). Gonadal hormones, such as oestrogen and progesterone, modulate dopamine activity and receptor levels in the striatum and nucleus accumbens in female rats. In contrast, oestrogen does not affect striatal dopamine release in male rats, and male hormones, such as testosterone, do not affect the brain dopamine reward circuit (Becker 1999;Becker et al. 2005). These factors support the selection of male rats in these experiments.
Behavioural experiments demonstrated that 70 % of rats subjected to the CMS paradigm exhibited a decrease in the consumption of palatable sucrose solution, which suggests that these rats were vulnerable to stress. The remaining 30 % of the stressed animals were stress resilient, and these animals are often rejected from further analysis, because they do not exhibit the "appropriate" behavioural response to the stress regimen (Bergström et al. 2008).
The stress-resilient group of animals is important for the identification of the molecular factors that underlie the mechanisms of active coping for stress and adversity. The present study demonstrated that stress strongly affected the mesolimbic dopamine circuit in stress-resilient and stresssusceptible animals. This observation is consistent with that of Krishnan et al. (2007) and Blugeot et al. (2011), who demonstrated that stress-resilient and stress-susceptible animals exhibit elevated serum corticosterone levels. The  stress-resilient group subjected to CMS exhibited a downregulation of dopamine D 2 receptors in all of the examined brain structures except the medial VTA. This effect was significant following 2 weeks of the CMS procedure when compared with control animals, which suggests an internalisation of the receptors in response to the enhanced dopamine release, because no down-regulation of D 2 receptor biosynthesis (i.e. mRNA) was observed. However, the extension of CMS for 3 additional weeks (5 weeks total) produced a different effect, which clearly supports active processes in stress-resilient animals. This group displayed strong neurobiological plasticity in dopamine D 2 receptor density and mRNA expression, in contrast to stress-reactive animals. These alterations buffered the influence of stress and thus normalised the mesolimbic dopamine circuit via the elevation of dopamine D 2 receptor biosynthesis to control levels in stress-resilient animals. This effect was not observed in stress-reactive rats after 5 weeks of CMS. The normalisation of the brain reward circuit in stress-resilient animals may be the result of elevated dopamine D 2 autoreceptor biosynthesis in the VTA region, which decreases the firing rate of dopamine neurons that project to the ventral striatum in response to stress. D 2 mRNA in the striatum and nucleus accumbens is responsible for the  Fig. 4 The effect of 2 weeks of chronic mild stress on dopamine D 2 receptor mRNA levels in various regions of the rat brain. a Lateral striatum; b medial striatum; c nucleus accumbens septi, core; d nucleus accumbens septi, shell; e lateral ventral tegmental area; f medial ventral tegmental area. Data represent the means ± SEM; n=7-10; ANOVA, nonsignificant biosynthesis of dopamine D 2 receptors in non-dopaminergic cells in these brain regions. We observed alterations in D 2 mRNA levels in the VTA and regions receiving projections from dopaminergic cells after 5 weeks of CMS. Therefore, we conclude that the biosynthesis of both pre-and post-synaptic dopaminergic D 2 receptors was increased. These results further support the hypothesis that stress resilience is a biochemically and biologically active and dynamic process that occurs very early at multiple levels in organisms that face trauma or adversity (Haglund et al. 2007;Kanarik et al. 2011). Moreover, stress-reactive animals may exhibit a delayed dopamine D 2 receptor response to stress when compared with stressresilient animals, which produces an anhedonic mood. However, this hypothesis must be tested in further studies.
Chronic and unpredictable stress regimens decrease dopamine D 2 mRNA expression and receptor density in the mesoaccumbens circuit in stress-reactive animals (Puglisi-Allegra et al. 1991;Papp et al. 1994;Dziedzicka-Wasylewska et al. 1997;Cabib et al. 1998;Zhu et al. 2010). Dopamine D 2 receptor density also increases after chronic antidepressant treatment, which supports the potential antidepressant efficacy of D 2 receptor activation (Gershon et al. 2007;Dunlop and Nemeroff 2007). Therefore, decreased D 2 receptor density in mesolimbic system may be a primary factor in reward subsensitivity and anhedonia, which is a core symptom of depression (Papp et al. 1994;Cabib et al. 1998). However, few studies have focused on the molecular, neurobiological and biochemical changes in dopamine D 2 signalling in stress-  resilient animals (Krishnan et al. 2007;Bergström et al. 2008;Kanarik et al. 2011;Blugeot et al. 2011). Bergström et al. (2008) have demonstrated that restraint stress and 5 weeks of CMS do not alter dopamine D 2 receptor mRNA expression in the nucleus accumbens of control, stress-reactive and stress-resilient rats. However, our work focused on the terminal and brain regions that contain cell bodies (medial and lateral VTA) of the dopamine mesolimbic system. Furthermore, we used a mixture of three oligonucleotides that are complementary to three different regions of the D 2 receptor mRNA to enhance the sensitivity of in situ hybridisation. The statistically significant parallel increases in dopamine D 2 mRNA expression and receptor density in the medial and lateral VTA in stress-resilient animals demonstrate that the mesolimbic circuit can effectively adapt to new circumstances and allow the animals to cope with adversity despite their sensitivity to chronic stress. In contrast, stressreactive rats exhibited a significant and persistent decrease in dopamine D 2 receptor density and unchanged mRNA expression levels. This phenomenon suggests that these animals exhibit an impaired mesoaccumbens reward circuit that cannot actively cope with stress (Stamford et al. 1991;Cabib et al. 1998;Kruk et al. 1998;Di Chiara et al. 1999;Cao et al. 2010).  Fig. 6 The effect of 5 weeks of chronic mild stress exposure dopamine D 2 receptor mRNA levels in various regions of the rat brain. a Lateral striatum; b medial striatum; c nucleus accumbens septi, core; d nucleus accumbens septi, shell; e lateral ventral tegmental area; f medial ventral tegmental area. Data represent the means ± SEM; n=10-11; ANOVA followed by Bonferroni's multiple comparison post hoc test; *p<0.05; **p<0.01; ***p <0.001 We used domperidone to visualise dopamine D 2 receptor density in various brain regions. This compound is a specific antagonist that recognises the high-and low-affinity conformational states of the D 2 receptor (Seeman et al. 2003(Seeman et al. , 2005. Therefore, the observed changes in [ 3 H]domperidone receptor binding in the VTA and NAcc in both groups of animals may be the sum of two phenomena: the down-regulation of D 2 dopamine receptors and the switching of their conformation into a low-affinity state (Rogoz and Dziedzicka-Wasylewska 1999;Seeman et al. 2005).
Our study suggests that the links between behavioural and biochemical responses to stress are not straightforward. The behavioural response to stress depends on various factors and systems and their mutual relationships at different levels (Manji et al. 2001). The behavioural CMS experiment suggests that resilient animals exhibit de facto resistance to stress, which is expressed as unaltered consumption of a 1 % sucrose solution. Biochemical analyses indicated an imbalance in mesolimbic dopamine circuit in stress-resilient rats as demonstrated by a decreased in [3H]domperidone binding when compared with stress-reactive group after 2 weeks of CMS. Mesolimbic D 2 receptor density in the stress-resilient group, but not the stress-reactive group, returned to normal levels after 5 weeks of CMS in all examined brain regions, except the NAcc core region, which is related to the caudate and putamen, and, in effect, less associated with the limbic reward system (Deutch and Cameron 1992). Stress influences dopamine D 2 receptors and other catecholamine circuits and these common fluctuations produce specific behavioural responses to stress that may not be directly related to changes in dopamine D 2 receptors alone. An examination of all of these factors is very important for a profound understanding on the mechanisms underlying stress resilience. These results clearly demonstrate that despite earlier blunting, the activation of dopamine receptor biosynthesis in the dopamine mesoaccumbens system in stress-resilient rats is involved in active coping for stressful experiences, while in stress-reactive animals, this phenomenon does not occur or may be delayed (Cabib and Puglisi-Allegra 1994;Puglisi-Allegra et al. 1991). However, further investigation is required to delineate the precise mechanisms underlying the time-dependent normalisation of dopamine D 2 receptor levels in the mesolimbic system.