[1] 14w Repeated paroxetine treatment reverses anhedonia induced in rats by chronic mild stress or dexamethasone

[1] 148w The present study was designed to assess the effect of dexamethasone, a synthetic glucocorticoid receptor agonist, in the sucrose preference test in rats. Rats treated acutely with dexamethasone (5-10 mg/kg) showed a significant decrease in sucrose preference (anhedonia) in comparison to vehicle treated rats, although 1 mg/kg dexamethasone did not alter the sucrose preference. Daily paroxetine treatment (10 g/kg, i.p., 14 days) reversed the anhedonic effect of acute dexamethasone (5 mg/kg), while causing no increased sucrose preference in rats that received dexamethasone vehicle. The paroxetine vehicle treated rats showed anhedonia even 14 days after acute dexamethasone administration. Paroxetine (10 mk/kg, i.p. for 28 days) also reversed anhedonia induced by chronic mild stress (8 weeks). In conclusion, acute dexamethasone induced an enduring anhedonic state that was reversed by repeated paroxetine treatment. Thus, the present study adds new data to the evidence supporting an important role for glucocorticoid in depression.

[1] 251w The precipitation of a depressive episode has been linked to stressful life events (Paykel, 2003), which can activate the hypothalamic-pituitary-adrenal (HPA) axis leading to an increase in plasma cortisol. Furthermore, dysregulation of HPA; such as lack of cortisol suppression by dexamethasone administration, increased cortisol secretion and blunted adrenocorticotropic hormone response to exogenous corticotropin-releasing hormone; are frequently associated with depression, suggesting that depression is related to a failure in HPA negative feedback, which would result in higher cortisol levels (Checkley, 1992;Holsboer, 2001;Barden, 2004;Juruena et al., 2004). This stress-induced increase in cortisol secretion is one underlying mechanism proposed for the stress-depression association (Holsboer, 2001;Mello et al., 2003;Paykel, 2003). The hippocampus, which shows signs of atrophy in patients with prolonged depression (Campbell and MacQueen, 2004), is vulnerable to stress and increased glucocorticoid levels (Campbell and MacQueen, 2004;McEwen, 2005). Since the hippocampus exerts a negative control over the hypothalamic-pituitary-adrenal (HPA) axis, its atrophy may induce impairment in HPA control, leading to cortisol hypersecretion (Holsboer, 2001;Barden, 2004). Thus, glucocorticoid is thought to play a major role in hippocampal atrophy and depressive symptoms. In this context, chronic antidepressant administration was shown to increase corticosteroid receptors, which can restore HPA negative feedback and normalize cortisol levels and HPA function (Barden, 2004). Furthermore, an abnormal dexamethasone suppression test after treatment-induced clinical improvement is associated with a higher risk of relapse and may present prognostic value for treatment (Dratcu and Calil, 1989;Ribeiro et al., 1993). Thus, it appears that there is an interrelationship between stress, high glucocorticoid levels and depression.

[2] 345w Several animal models of depression, such as the forced swim test, learned helplessness and anhedonia induced by unpredictable chronic mild stress, involve behavioral responses to stressful procedures. The anhedonia induced by unpredictable chronic mild stress consists in the repeated exposure of animals to unpredictable mild stress: tilted cage, food deprivation, paired caging, reduction of cage area, etc.; leading to a reduction in self-stimulation of rewarding areas or in the consumption of palatable food or liquids, i.e. anhedonia. This model yields good similarity with clinical depression, since it was found that stressful life events frequently precede major depression episodes (Paykel, 2003;Willner, 2005). This model has also good face validity, since anhedonia, described as a marked diminished interest or pleasure in events that would normally be enjoyable, is a core symptom of major depression episodes according to DSM-IV criteria (American Psychiatric Association, 1994;Willner, 2005). Furthermore, the reversal of anhedonia induced by unpredictable chronic mild stress requires 2-4 weeks of daily antidepressant treatment (Willner, 1997;Stout et al., 2000;Willner, 2005), again showing good parallels with clinical data. Therefore, the chronic mild stress paradigm is adequate for providing insight into the neurobiology of depression (Willner, 1997(Willner, , 2005)). However, the anhedonia induced by unpredictable chronic mild stress was not reliably observed in some experiments (Harris et al., 1997;Nielsen et al., 2000;Bielajew et al., 2002). Another problem is that although some studies found an increase in plasma corticosterone in this model (Ayensu et al., 1995;Harris et al., 1997;Bielajew et al., 2002;Froger et al., 2004;Grippo et al., 2005a;Song et al., 2006), this effect was not consistently observed (Harris et al., 1997;Stout et al., 2000;Grippo et al., 2005b). Although these inconsistencies could be related to the strain of the experimental animals or procedural differences (e.g. nature, duration or frequency of stress) among the studies, they may be also related to individual variability and styles of coping with stress (Nielsen et al., 2000;Bielajew et al., 2002;Veenema et al., 2003;Anisman and Matheson, 2005). The administration of exogenous glucocorticoid would avoid some of these variables in the study of the role of glucocorticoid in stress-induced depression.

[3] 79w Thus, the main objective of the present study was to evaluate the effect of acute administration of dexamethasone, a synthetic glucocorticoid that binds to glucocorticoid receptors, on the sucrose preference of rats; a measure of anhedonia. If the dexamethasone-effect on anhedonia plays a significant role in depression neurobiology, it should be reversed by repeated antidepressant administration. Therefore, the influence of chronic treatment with paroxetine, a clinically effective antidepressant drug, regarding the effect of dexamethasone on anhedonia was also studied.

[1] 116w The rats were initially divided into two groups: stressed and nonstressed. The stressed group received a stress regimen over an eight-week period, consisting of weekly "unpredictable" (in fact, a pseudorandom sequence) mild stress, such as food and/or water deprivation, an overnight cage tilt, an overnight soiled cage, space reduction and continuous overnight illumination (Table 1). The pseudorandom sequence of stressors was used to avoid the rats developing any habituation to repeated mild stress. The nonstressed (control) group remained undisturbed, except for the previously described deprivation before the weekly sucrose consumption (each Tuesday, between 8:00 and 10:00 am) test and the handling necessary for animal care (cleaning cages) and drug administration (weighing, tail marking and drug administration).

[2] 59w After 4 weeks the stressed and nonstressed groups were subdivided into two paired subgroups (see above, n = 6-8 rats/ group): (1) stressed administered 10 mg/kg paroxetine; (2) stressed administered distilled water as vehicle; (3) nonstressed administered paroxetine and; (4) nonstressed administered vehicle. Both treatments were administered for 28 days, intraperitoneally (i.p.), at a constant volume of 1.0 ml/kg.

[3] 105w Two-way ANOVA indicated a significant effect for treatment: [F (3, 16) = 5.34, p b 0.0019]; weeks [F(2, 32) = 11.21, p b 0.0001] and interaction [F(6, 32) = 2.44, p b 0.0020]. As shown in Fig. 2, significant group differences occurred after 24 h [F(3, 16) = 5.98, p b 0.01] and 48 h [F(3, 16) = 7.09, pb 0.01] of dexamethasone administration. On both occasions a dose-dependent anhedonia induced by dexamethasone occurred, with a significant difference between the 5 and 10 mg/kg dexamethasone groups from the control group (all pb 0.01). No significant difference was seen in baseline [F(3, 160) = 1.24, NS].

[4] 62w Significant reductions also occurred in sucrose preference from baseline (day 0 of the same group) in rats treated with dexamethasone 5 mg/kg [F(2, 8) = 11.37, pb 0.01] and 10 mg/kg [F(2, 8) = 28.26, pb 0.001]. The group treated with 1 mg/kg dexamethasone exhibited no difference from the control group at 24 and 48 h or from baseline [F(2, 8)=1.25, NS].

[5] 63w The anhedonia exhibited by the 5 and 10 mg/kg dexamethasone groups was stable for 1 week (data not shown), when the experiment was finished. However, 60% of the rats that were administered 10 mg/kg dexamethasone died. Since a 5 mg/kg dexamethasone dose was able to induce anhedonia without mortality or other external signs of toxicity, this dose was chosen to perform subsequent experiments.

[6] 136w Dexamethasone administration presented a significant effect on the total fluid intake throughout the experiment [F(92, 32) = 30.22, p b 0.001] and week× treatment interaction [F(6, 32)= 2.89, p b 0.03], but not for treatment [F(3, 16) = 2.56, pN 0.05]. However, no significant difference was seen among treatments in any sucrose preference test [baseline: F(3, 16) = 2.83; 24 h: F(3, 16)= 2.64; 48 h: F(3, 16) = 2.47; all pN 0.05]. their baselines (p b 0.001). Seven days of paroxetine treatment (day 9) was unable to modify this result (p b 0.001), but 14 days of paroxetine treatment (day 16) reversed the anhedonic-like state induced by dexamethasone. However, the dexamethasone plus distilled water treated group exhibited a persistent anhedonia even 16 days after dexamethasone treatment (day 14 of distilled water treatment; p b 0.001).

[7] 93w In the CMS procedure (experiment 1), a significant effect occurred in body weight gain [nonstressed + water: 71.4± 9.3 g; nonstressed + paroxetine: 51.3± 7.4 g; stressed+ water: -21.4 ± 11.6 g; stressed+ paroxetine: 5.0 ± 9.4 g; mean ± SEM; F(3, 25)= 19.13, pb 0.0001], which resulted from a higher weight gain in nonstressed rats (saline or paroxetine treated) than stressed saline treated rats (both pb 0.001). Moreover, observation also revealed a reduced weight gain in the stressed group treated with paroxetine when compared to nonstressed saline treated rats (both pb 0.01).

[8] 58w In dexamethasone-induced anhedonia (experiment 2), all the dexamethasone treated groups exhibited a significant reduction in body weight gain [vehicle: 24.6 ±2.1 g; DEX 1 mg/kg: 3.8 ± 7.2 g; DEX 5 mg/kg: -6.0 ±6.9 g; DEX 10 mg/kg: -13.4 ± 3.2 g; mean± SEM; F(3, 16) = 9.44, pb 0.001] compared to vehicle treated rats (all pb 0.02).

[9] 96w In experiment 3 (paroxetine effect on dexamethasoneinduced anhedonia), a significant effect on body weight gain also occurred [vehicle + water: 24.0 ± 2.5 g; vehicle + paroxetine: 19.0 ± 5.8 g; DEX + water: -7.0 ±14.8 g; DEX + paroxetine: 1.0 ± 7.9; mean± SEM; F(3, 16)= 5.75, pb 0.01]. Post-hoc comparison showed that the vehicle plus vehicle group presented a higher body weight gain than both dexamethasone treated groups (plus saline or paroxetine; both pb 0.05). Furthermore, the vehicle plus paroxetine group presented a higher body weight gain than dexamethasone plus distilled water (pb 0.05).

[1] 99w The main finding of the present study is that dexamethasone induced a decrease in sucrose preference, which is suggestive of an anhedonic state. This anhedonia was similar to that found in the chronic mild stress model. Additionally, repeated, but not acute, administration of paroxetine (a clinically effective antidepressant drug) gradually restored the sucrose preference, indicating that paroxetine was able to reverse the dexamethasone-induced anhedonia, showing a similar profile to repeated paroxetine administration on anhedonia induced by chronic mild stress. Since no significant difference in total fluid consumption between the groups was observed, these effects were specific to sucrose preference.

[2] 284w The fact that anhedonia is one of the core criteria for major depression diagnosis and that it can be induced in rats by dexamethasone administration, supports the view that corticosteroid plays an important role in the neurobiology of depression. However, in contrast to the dexamethasoneinduced anhedonia found in the present study, corticosteroid is also associated with rewarding effects. For example, chronic corticosterone administration reduced the current threshold for hypothalamic self-stimulation (Barr et al., 2000). This discrepancy may be related to methodological differences, like the difference in sensitivity of the hedonic indices (Nielsen et al., 2000) and corticosteroid administration; acute dexamethasone or chronic corticosterone administration. On the other hand, the anhedonic effect of dexamethasone observed in the present study is in agreement with other data, which found that high corticosteroid administration reduces sexual performance (Gorzalka and Hanson, 1998), a behavioral change that also is included in the anhedonia criteria for a major depressive episode (American Psychiatric Association, 2000). It is interesting that in depressive patients, failure in the dexamethasone suppression test was associated with anhedonia and suicide ideation (Oei et al., 1990). Furthermore, antidepressant treatment can normalize hypothalamic-pituitary-adrenal dysfunctions in patients with major depression and increase brain corticosteroid receptors in both animals and humans (Holsboer, 2001;Calfa et al., 2003;Barden, 2004;Juruena et al., 2004). Reinforcing the glucocorticoid role in depression, it has been shown that strategies that reduced glucocorticoid effects exerted an antidepressant-like effect in animal models (Veldhuis et al., 1985;De Kloet et al., 1988;Mitchell and Meaney, 1991;Peeters et al., 1992;Papolos et al., 1993;Baez and Volosin, 1994;Peeters and Broekkamp, 1994;Korte et al., 1996;Bachmann et al., 2005;Gregus et al., 2005;Rogoz et al., 2005;Johnson et al., 2006) and an antidepressant effect in patients (Young, 2006;Berton and Nestler, 2006).

[3] 502w The acute administration of dexamethasone induced an anhedonic state 24 h after treatment and this effect persisted until the end of the study; 16 days after dexamethasone administration. Thus this effect was not due to the presence of the drug, but the consequence of a more stable alteration induced by glucocorticoid. One structure that may mediate this deleterious effect of high glucocorticoid levels is the hippocampus, since a reduction in its volume has been frequently associated with depression and it is sensitive to high glucocorticoid levels (Graeff et al., 1996;Haynes et al., 2001, 2004;Campbell and MacQueen, 2004;McEwen, 2005). Considering the glucocorticoid receptor activity, the dexamethasone doses used in the present study were high when compared directly to a corticosterone dose used to mimic its rise in stressful procedures (Andreatini and Leite, 1994;Retana-Marquez et al., 2003), to block neurogenesis (Hellsten et al., 2002;Mayer et al., 2006), to increase immobility time (Johnson et al., 2006) or to reinstate the immobility time of adrenalectomized rats in the forced swimming test (Jefferys et al., 1983;Veldhuis et al., 1985;Mitchell and Meaney, 1991;Peeters et al., 1992;Peeters and Broekkamp, 1994). However, it is within the dose range of DEX used to induce hippocampal damage; 0.7 to 20 mg/kg (Haynes et al., 2001(Haynes et al., , 2004)). Interestingly, this dexamethasone-induced hippocampal damage was attenuate by chronic pretreatment with antidepressants of different classes (Haynes et al., 2004). Moreover, factors other than receptor potency (e.g. multidrug-resistant P-glycoprotein on the blood brain barrier) may contribute to the magnitude of the glucocorticoid effect on the brain (Buckingham, 2006), which could partially explain this discrepancy between the dexamethasone and corticosterone dose range. In future experiments, it will be interesting to evaluate the effect of repeated low doses of corticosterone or dexamethasone on the sucrose preference test. The cellular mechanisms through which dexamethasone induces such effects were not addressed in the present study, but these effects may be mediated by brain derived neurotrophic factor, since glucocorticoid and stress decrease this factor and such effects are reversed by chronic antidepressant treatment (Dwivedi et al., 2006). Similarly, electroconvulsive seizures found in an animal model of electroconvulsotherapy in humans, reverse the reduction of neurogenesis induced by corticosterone administration (Hellsten et al., 2002). The time for paroxetine reversal of dexamethasone-induced anhedonia (14 days) is shorter than for CMS-induced anhedonia (21 days). This may be due to the fact that in dexamethasone-induced anhedonia, no other DEX administration occurred during the treatment with paroxetine, thus the rats were exposed only to one episode of high glucocorticoid plasma level; while in CMSinduced anhedonia, the stressful events remain during the drug treatment, thus these rats were submitted to repeated episodes of increased glucocorticoid plasma levels. Another possibility that could explain this difference is interexperiment variability. On the other hand, the prolonged anhedonic-state of at least 2 weeks induced by one high dexamethasone dose administration showed that concurrent high glucocorticoid levels and depressive symptoms are not a requirement, which may partially explain the fact that high cortisol levels are not found in all depressed patients.

[4] 130w In view of the fact that certain reproducibility problems with anhedonia induced by pseudorandomized chronic mild stress present some difficulties with the use of this model (Vollmayr and Henn, 2003;Willner, 1997), the results of experiment 1 showed that the procedure was able to induce anhedonia, under the experimental conditions studied. Moreover, repeated, but not acute, paroxetine treatment reversed this anhedonia, with no alteration in total fluid intake. Although the reversal of CMS-induced anhedonia has already been shown with other SSRI, like fluoxetine, sertraline, citalopram and escitalopram (Muscat et al., 1992;Marona-Lewicka and Nichols, 1997;Przegalinski et al., 1995;Montgomery et al., 2001) to our knowledge, this is the first report showing the ability of chronic paroxetine treatment to reverse anhedonia induced by chronic mild stress, which increases the pharmacological validity of the model.

[5] 298w Since both CMS (Matthews et al., 1995) and high glucocorticoid administration (Andreatini and Leite, 1994) can lead to body weight change and that controversy exists regarding the influence of body weight change on sucrose solution consumption (Matthews et al., 1995;Willner et al., 1996), it is important to measure this variable. In the present study, the CMS procedure led to lower body weight gain compared to saline treated nonstressed animals. All dexamethasone doses also led to reduced body weight gain compared to vehicle treated rats. In the last experiment regarding paroxetine reversal of DEX-induced anhedonia, both DEX treated groups also showed lower body weight gain compared to vehicle treated rats. These data are in agreement with previous data showing that chronic mild stress and high glucocorticoid administration lead to reduced body weight gain, which could be viewed as a confounding variable in the anhedonia results and, thus, could lead to false results. However, some points may indicate that this is not the case in the present study. First, it was suggested that although absolute sucrose consumption may be influenced by body weight gain, sucrose preference may not be influenced, which makes this latter parameter a better index of anhedonia (Matthews et al., 1995). Second, the CMS and DEX administration induced a reduced weight gain both in water and paroxetine treated rats, although only the former showed a decrease in sucrose preference at the end of experiments. These results suggest some dissociation between body weight change and anhedonia. Finally, it is important to note that body weight reduction is also a diagnostic criterion for a major depressive episode in DSM-IV (American Psychiatric Association, 1994) and their occurrence in the CMS procedure may be considered another indication of the face validity of the CMS procedure, instead of only a confounding variable.

[6] 86w In conclusion, the present results suggest that high glucocorticoid levels induce an anhedonic state, which can be reversed by repeated antidepressant treatment. These results reinforce the hypothesis that glucocorticoids play an important role in stress-induced depression and indicate that high glucocorticoid levels can lead to depression and not the opposite. However, depression is a multifactorial disorder and hypothalamic-pituitary-adrenal dysfunction must be viewed as only one of many contributing factors. Additionally, the procedure described here is a useful approach for studying the role of glucocorticoid in depression.

[1] 159w Adult male Wistar rats weighing between 200 and 300 g were used. The rats were housed in polypropylene cages with wood shavings as bedding, under controlled room conditions of light (12-h light-dark cycle, lights on at 7:00 a.m.) and temperature (22 ± 2 °C), with free access to food and water, except prior to the sucrose preference test or when they were submitted to chronic mild stress (see below). Two rats were housed in each cage (cage size: 41 × 32 × 16.5 cm) but they were isolated by a central aluminum wall, which divided the cage in two equal compartments and permitted minimal contact between them, but neither one consumed the food or water/sucrose solution of the other. Thus, the rats were not absolutely isolated. All procedures were carried out in compliance with the NIH Guide for the Care and Use of Laboratory Animals (Committee to Revise the Guide for the Care and Use of Laboratory Animals, 1996).

[2] 93w Paroxetine (Eurofarma, São Paulo, Brazil) was dissolved in distilled water. Dexamethasone-acetate (DEG, Curitiba, Brazil) was suspended in saline containing 0.2% Tween 80. The vehicle of each drug was administered in the respective control rats. All drugs were administered intraperitoneally (i.p.) at a constant volume of 1.0 ml/kg. Dexamethasone or its vehicle was administered once, while paroxetine or its vehicle was administered for 14 (dexamethasone-induced anhedonia) or 28 days (chronic mild stress experiment). The paroxetine dose was chosen on the basis of previous studies in our laboratory (Consoni et al., 2006;Beijamini and Andreatini, 2003).

[3] 258w In all experiments, prior to the first sucrose preference test, all the rats were submitted to 48 h of forced exposure to 1% sucrose solution in order to habituate to it, during which sucrose solution was the only fluid available for consumption, followed by two days of free access to food and water. After this, the rats were submitted to water deprivation for 16 h prior to performing the sucrose preference test; baseline test at day zero. The sucrose preference test was performed in the rat's home cage: two pre-weighted bottles, one containing tap water and another containing 1% sucrose solution, were presented to each rat. The bottles were weighed again after 1 h and the weight difference was considered to be the rat intake from each bottle. The sum of water and sucrose intake was defined as total intake and the sucrose preference was expressed as the percentage of sucrose intake from the total intake following the formula: % sucrose preference ¼ sucrose intake  100= total intake All tests were carried out weekly (each Tuesday) between 8:00 and 10:00 am, with a variable sequence of bottle positioning (for each rat, the side of sucrose or water bottles were changed from one test to another), in order to avoid habituation. After the sucrose preference test, all the rats received free access to food and water. After the baseline sucrose preference test, and prior to drug treatment or stress administration, the rats were paired according their preference and then distributed in experimental groups to form paired (matched) groups.

[4] 18w The body weight gain was calculated as the difference between the final and baseline (day 0) body weight.

[5] 29w Each experiment of the sucrose preference test was submitted to two-way ANOVA with repeated measures, with drug treatment as an independent factor and treatment weeks as a dependent factor.

[6] 307w Whenever a significant treatment × trial interaction was found, intergroup comparisons were realized for each week using one-way ANOVA, followed by the Newmann-Keuls post-hoc test. Intragroup comparison was carried out by repeated measures ANOVA followed by the Newmann-Keuls post-hoc test. Body weight gain was analyzed by one-way ANOVA followed by the Newman-Keuls posthoc test. Statistical significance was considered when p b 0.05. 18:00----10:00 18:00----10:00 Continuous overnight illumination 18:00---10:00 18:00---10:00 Space reduction 18:00----10:00 Soiled cage 18:00--8:00 18:00---10:00 Water deprivation 18:00------18:00 Food deprivation 18:00-------18:00 Food + water deprivation 10:00------10:00 p b 0.001] and day 56 [F(3, 25) = 11.19, p b 0.001]. No significant effect was seen between baseline [day 0: F(3, 25) = 0.07, NS] and day 7 [F(3, 25) = 0.78, NS]. Prior to paroxetine treatment onset (day 28), a significant reduction in sucrose preference (anhedonia) occurred in both stressed groups in comparison to nonstressed groups on the same day (both p b 0.01), an effect that was sustained until day 42 for paroxetine (all p b 0.01) treated rats and until day 56 for water treated rats (all p b 0.05). On day 56 the stressed group treated with paroxetine showed no difference in comparison with the nonstressed group, but differed significantly from stressed rats treated with vehicle (p b 0.001). Surprisingly, a significant effect occurred between the stressed plus paroxetine and stressed plus water groups on day 14 and 21 (both p b 0.02). Furthermore, a significant difference occurred between nonstressed rats and stressed plus water rats on day 21 (both p b 0.001). A difference in sucrose preference occurred throughout the experiment in all groups (nonstressed+ water: F(8, 40) = 2.80, p b 0.02; nonstressed + paroxetine: F(8, 56) = 7.26, p b 0.001; stressed + water: F(8, 48) = 3.81, p b 0.01; stressed + paroxetine: F(8, 56) = 11.47, p b 0.001).

[1] 63w This experiment was performed to assess the possible role of glucocorticoid in anhedonia in rats. The rats (n = 20) were submitted to a baseline sucrose preference test and then allocated to one of four paired groups: vehicle (control), 1, 5, and 10 mg/kg dexamethasone. Twenty-four and forty-eight hours later, respectively, the rats were submitted to a second and third sucrose preference test.

[2] 114w In this experiment, after a baseline sucrose preference test, 20 rats were divided into two paired groups (n = 10/group), receiving a single dose of dexamethasone or vehicle. Then, following a second sucrose preference test (48 h), these groups were further subdivided into two paired groups (n = 5/group), one treated with distilled water and the other with 10 mg/kg paroxetine. Thus, 4 groups were formed: (1) vehicle plus distilled water; (2) vehicle plus 10 mg/kg paroxetine; (3) dexamethasone plus distilled water; and (4) DEX plus 10 mg/kg paroxetine. On the experimental day, the paroxetine administration occurred 2 h after the sucrose preference test, which occurred weekly, each Tuesday, between 8:00 and 10:00 am.

[3] 104w Post-hoc analysis in both nonstressed groups showed that no significant difference occurred between baseline and any other day. In contrast, a significant reduction occurred in sucrose preference on some day when compared to baseline results in both stressed groups. In stressed rats treated with water, the sucrose preference was significantly lower that baseline on days 14, 21, 24, 28, 49 and 56 (all p b 0.03). On day 35 and 42 the differences were almost significant (both p = 0.06). In stressed rats treated with paroxetine, the sucrose preference was significantly lower than baseline on day 14, 21 and 28 (all p b 0.05).

[4] 18w The CMS procedure led to a significant change in the total fluid intake throughout the experiment [F( 8

[5] 13w No significant effect was seen in total fluid intake for treatment [F(3, 16)