Chronic stress effects on working memory: Association with prefrontal cortical tyrosine hydroxylase

Chronic stress effects on working memory: Association with prefrontal cortical tyrosine hydroxylase

Behavioural Brain Research 286 (2015) 122–127 Contents lists available at ScienceDirect Behavioural Brain Research journal homepage: www.elsevier.co...

1MB Sizes 0 Downloads 35 Views

Behavioural Brain Research 286 (2015) 122–127

Contents lists available at ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Short Communication

Chronic stress effects on working memory: Association with prefrontal cortical tyrosine hydroxylase Young-A Lee 1 , Yukiori Goto ∗,2 Department of Psychiatry, McGill University, Montreal, QC, Canada

h i g h l i g h t s • Chronic stress affects working memory function in rats. • Altered working memory is associated with prefrontal cortical tyrosine hydroxylase. • Chronic stress effects on working memory depend on prefrontal tyrosine hydroxylase.

a r t i c l e

i n f o

Article history: Received 4 January 2015 Received in revised form 22 February 2015 Accepted 2 March 2015 Available online 6 March 2015 Keywords: Dopamine Working memory Stress Prefrontal cortex Tyrosine hydroxylase Psychiatric disorder

a b s t r a c t Chronic stress causes deficits in cognitive function including working memory, for which transmission of such catecholamines as dopamine and noradrenaline transmission in the prefrontal cortex (PFC) are crucial. Since catecholamine synthesis depends on the rate-limiting enzyme, tyrosine hydroxylase (TH), TH is thought to play an important role in PFC function. In this study, we found that two distinct population existed in Sprague-Dawley rats in terms of working memory capacity, one with higher working memory capacity, and the other with low capacity. This distinction of working memory capacity became apparent after rats were exposed to chronic stress. In addition, such working memory capacity and alterations of working memory function by chronic stress were associated with TH expression in the PFC. © 2015 Elsevier B.V. All rights reserved.

The prefrontal cortex (PFC) is one of the central brain regions that mediate cognitive function [1]. The PFC receives catecholamine inputs from the ventral tegmental area (VTA) and locus coeruleus (LC), where dopamine (DA) and noradrenaline (NA) neurons, respectively, are located [2,3]. These catecholamines in the PFC are crucial for such cognitive function as working memory [2,3]. In this regard, tyrosine hydroxylase (TH), which is the rate-limiting enzyme for catecholamine synthesis ([4]), is thought to play a critical role in PFC function. TH is expressed not only in DA and NA cell bodies, but also localized in varicosities of catecholamine axonal fibers that project into the PFC, suggesting that DA and NA are locally synthesized in the PFC.

∗ Corresponding author at: Yukiori Goto, PhD, Primate Research Institute, Kyoto University, 41-2 Kanrin, Inuyama, Aichi, 484-8506, Japan Tel.: +81 568 63 0551; fax: +81 568 63 0551. E-mail address: [email protected] (Y. Goto). 1 Present address: Department of Food Science & Nutrition, Catholic University of Daegu, Gyeongsan, Gyeongsangbuk-do, South Korea. 2 Present address: Primate Research Institute, Kyoto University, Inuyama, Aichi, Japan. http://dx.doi.org/10.1016/j.bbr.2015.03.007 0166-4328/© 2015 Elsevier B.V. All rights reserved.

Chronic stress yields devastating effects including cognitive dysfunction [5]. In rodents, working memory has been reported to be disrupted by chronic stress [6]. Studies with microdialysis analyses have revealed that DA transmission is decreased in the PFC of rodents exposed to chronic stress [6,7]. In contrast, chronic stress has been also shown to increase TH expression in VTA DA [8] and LC NA neuron cell bodies [9]. It has remained unknown whether TH expression in axons of catecholamine neurons projecting into the PFC is affected by chronic stress. In particular, working memory has been shown to depend on both DA and NA transmission in the PFC [2,3]. Moreover, chronic stress also affects both DA [6,7] and NA [7] release in the PFC. Collectively, the effects of chronic stress on working memory could be produced by combined alterations of both DA and NA, but neither DA or NA alone. Thus, TH appears to be a promising candidate that is influenced by chronic stress, as this enzyme is involved in synthesis of both DA and NA. In this study, we examined whether TH expression in the PFC of rats was associated with working memory, and whether such association was modulated by chronic stress. All experiments were conducted in accordance with the Canadian Council on Animal Care Guide to the Care and Use of Experimental

Y.-A. Lee, Y. Goto / Behavioural Brain Research 286 (2015) 122–127

Animals, and approved by the McGill University. Male adult Sprague-Dawley rats were purchased from Charles River Laboratories (St-Constant, QC, Canada). Chronic stress was given to rats by either placing them in a clear acrylic restrainer for 6 h per day for 21 days (i.e. chronic restraint stress), or placing them on a clear transparent acrylic platform (20 × 20 cm) elevated 1 m from the ground floor for 1 h per day for 14 days (i.e. chronic elevated platform stress). These chronic stress procedures for rodents have been utilized and well-documented in previous studies [10–16]. A spatial delayed alternation (SDA) task was conducted using the T-maze (Fig. 1A) to examine chronic stress effects on working memory [6,17]. In this task, animals had to learn a rule to turn into the left and right arm of the maze alternately to obtain rewards (cereals). Before training, rats were habituated to the maze during which they freely explored the maze for 20 min per day for 3 days. Rats were then trained to perform the task. Rats received training sessions, consisting of 24 trials per session per day, until they performed over 80% correct responses per session for three consecutive sessions. No inter-trial interval (ITI) was incurred in these training sessions. When animals made incorrect responses (i.e. entering into the same arm consecutively), additional correction trials were given until they made a correct response. Once rats completed training, three test sessions were conducted, in which ITIs were set at 0, 10, and 30 s, respectively, in each session. A percentage of correct responses (%CR) in each test session was recorded. Throughout the task, food restriction was given to animals, during which body weights of both stressed and control rats were carefully monitored, and maintained not to be lower than 85% of body weights of rats with ad libitum food access. In the SDA task, 29 rats were tested. These animals were divided into those exposed to (1) restraint stress (n = 13), (2) elevated platform stress (n = 10), and (3) without stress (n = 6). Rats were first trained to perform the SDA task, and then subjected for stress exposure (Fig. 1A). The SDA task was conducted again following stress exposure (started one day after the last stress exposure). Performance before and after stress exposure were recorded. All data are expressed as means ± s.e.m. Statistical analysis was conducted using two-way repeated measures analysis of variance (ANOVA), with stress exposure (pre- vs. post-stress exposure) and delay conditions (0, 10, and 30 s delays) as within-subjects factors. Post hoc Tukey test was conducted for comparison of each group in ANOVA. We also examined TH expression in the PFC using immunohistochemistry. On a next day of the last stress exposure or the last test session of the SDA task, rats were euthanized with a lethal dose of sodium pentoberbital (100 mg/kg, i.p.), and transcardially perfused with ice-cold 0.1 M phosphate buffer saline followed by 4% paraformaldehyde. Brains were removed from the skulls, and post-fixed with 4% paraformaldehyde. These brains were cryoprotected with 30% sucrose solution, and cut into 40 ␮m sections with the microtome. The sections were washed and incubated with 0.5% H2 O2 to remove endogenous peroxides. Then, they were incubated with 2.5% horse serum to block non-specific binding of antibody. A primary antibody for TH (diluted at 1:1,000; the catalog number AB152; Millipore, Billerica, MA, USA) was incubated with the sections overnight at 4 ◦ C. The sections were further incubated with secondary biotinylated goat anti-rabbit IgG (1:1,000; the catalog number ab6721; Abcam, Cambridge, MA, USA) for 2 h followed by avidin-biotin-peroxidase complex for another 2 h, and visualized by reaction with 3 3-diaminobenzidine (DAB; Sigma–Aldrich, Oakville, ON). DAB reaction on the sections were further intensified by the silver-gold intensification method, for which the sections were incubated in 1% silver nitrate for 1 h at 55 ◦ C followed by 0.1% gold chloride for 10 min, and then 5% sodium thiosulfate for another 10 min at the room temperature. The sections were mounted on slide glasses, and hydrated by ethanol and xylene.

123

Antibody bindings against TH varicosities were examined using the light microscope. Images were acquired and stored into the computer through the CCD camera connected to the microscope for later off-line analysis of images. Using the particle analysis function of the ImageJ software (National Institutes of Health, USA), dot-like TH immunostaining that corresponded varicosities on catecholamine fibers in the superficial (layer II–III) and deep (layer V–VI) layers of the prelimbic (PL) and infralimbic (IL) cortex were quantified. Such quantification of varicosities resulted in more accurate TH immunostaining measurements than optodensitometric analysis, as it could exclude background artifact (e.g. blood vessels) staining. Twenty-two rats, which were a different bunch of rats from those 29 rats used in the first SDA task, were used for TH immunohistochemistry. Among these 22 rats, 10 rats were exposed to elevated platform stress, and the other 12 rats were stress-naive, controls. Rats were first subjected to perform the SDA task to examine a working memory capacity, which was expressed as the difference of %CR between the 0 and 30 s delay conditions. Such a working memory capacity was subsequently correlated with quantification of TH immunostatining in each rat. Statistical analysis was conducted using unpaired t-test or two-way ANOVA with stress (chronic stress vs. non-stress) and sub-regions within the PFC (PL vs. IL + superficial vs. deep layers) as independent variables. Rats before chronic stress exposure exhibited delay-dependent performance, with %CR getting lower as longer the delays were (n = 13; Fig. 1B). After three weeks of chronic restraint stress, these rats exhibited significantly lower %CR than that before stress exposure at the 0 s delay (stress, F1,36 = 5.62, p = 0.021; delay, F2,36 = 5.87, p = 0.005; stress × delay, F2,36 = 1.44, p = 0.245; post hoc test, p = 0.007 between pre- vs. post-stress at the 0 s delay), whereas %CR was not different between before and after stress exposure at the 10 and 30 s delays (Fig. 1B). This worsened performance after chronic stress exposure at the 0 s delay was not due to sub-optimal performance, since %CR at the 0 s delay in stress-naive rats (n = 6), which were trained for the SDA task and subjected for the task again 3 weeks after a blank period, was comparable between before and after the blank period (paired t-test comparing before and after the blank period, p = 0.61 at the 0 s delay; p = 0.50 at the 10 s delay; p = 0.47 at the 30 s delay; Fig. 1B). Moreover, motivation for food rewards in stressed rats was not decreased compared to that of control rats, as indicated by comparable engagement and performance of the SDA task at the longer delay (10 and 30 s) conditions between stressed and control rats. A similar alteration pattern on performance of the SDA task was also observed in rats exposed to chronic elevated platform stress (i.e. significant %CR decrease at the 0 s, but not other delay conditions after stress exposure; n = 10; stress, F1,27 = 3.48, p = 0.066; delay, F2,27 = 4.56, p = 0.018; stress × delay, F2,27 = 1.66, p = 0.199; post hoc test, p = 0.023 between pre- vs. poststress at the 0 s delay; Fig. 1 C). Therefore, rats with restraint and elevated platform stress were combined for further analysis. These results were unexpected, as a working memory deficit based on previous studies [6,17,18] was expressed as a greater impairment of performance in a longer delay than that in a shorter or no delay condition. Therefore, we investigated further details of the chronic stress-induced alterations of SDA task performance. Further analysis unveiled a bi-modal distribution of SDA task performance in chronically stressed rats (n = 23, rats with restraint and elevated platform stress; Fig. 1D). In one group of rats (n = 13; denoted “Group 1” hereafter), %CR at the 30 s delay was significantly lower than that at the 0 s delay (Fig. 1D). In contrast, rats in the other group (n = 10; denoted “Group 2” hereafter) exhibited lower %CR at the 0 s delay than that at the 30 s delay (Fig. 1D). In the Group 1, chronic stress worsened performance at the 30 s delay (stress, F1,36 = 9.35, p = 0.003; delay, F2,36 = 24. 6, p < 0.001; stress × delay, F2,36 = 1.97, p = 0.148; post hoc test, p = 0.002 between pre- vs.

124

Y.-A. Lee, Y. Goto / Behavioural Brain Research 286 (2015) 122–127

Fig. 1. Chronic stress-induced alterations of working memory in the SDA task. (A) A schematic diagram of the SDA task using the T-maze (left) and the experimental timeline (right). (B) A graph showing the effects of chronic restraint stress on performance of the SDA task. “Before stress” condition in the graph shows performance of the task that animals with stress exposure and no stress are combined. *p = 0.007, vs. pre-stress at the 0 s delay. Error bars indicate s.e.m. (C) A graph showing the effects of chronic elevated platform stress on performance of the SDA task. *p = 0.023, vs. pre-stress at the 0 s delay. (D) A histogram showing performance of the SDA task with chronic stress, expressed as difference of percentages of correct responses between the 0 and 30 s delays. A gray line in the graph indicates bimodal Gaussian distribution fitting. (E, F) Graphs showing performance of the SDA task before and after chronic stress exposure in the Group 1 (E) and Group 2 (F) that are denoted in Fig. 1D. *p = 0.002, vs. pre-stress at the 30 s delay, **p = 0.003, ***p = 0.007, vs. pre-stress at the 0 and 30 s delays, respectively. (G) A graph showing performance of the SDA task in the Group 1 and Group 2 before chronic stress exposure.

post-stress at the 30 s delay; Fig. 1E). In the Group 2, chronic stress impaired performance at the 0 s delay, but improved performance at the 30 s delay (stress, F1,27 = 1.37, p = 0.246; delay, F2,27 = 1.20, p = 0.306; stress × delay, F2,27 = 16.9, p < 0.001; post hoc test, p = 0.003 and p = 0.007 at the 0 and 30 s delays, respectively; Fig. 1F). It turned out that these distinct alterations with chronic stress between the Group 1 and Group 2 were also associated with their performance before stress exposure. Thus, %CR of rats before stress exposure was not statistically significant between the 0 and 30 s delays in Group 1, whereas %CR of Group 2 was significantly declined from the 0 to 30 s delays (unpaired t-test, p = 0.008; Fig. 1G). Since chronic stress has been shown to alter DA [6,7] and NA [7] release in the PFC, we examined TH expression in the PFC with immunohistochemistry. Varicosities on catecholamine axonal fibers were identified with TH staining (Fig. 2A). First, correlations between performance of the SDA task expressed as difference of %CR between the 0 and 30 s delays and TH expression in the PFC of stress-naive rats were examined. A significant correlation was observed between the density of varicosities in the PL superficial layer and working memory capacity (n = 12; r = −0.74, p = 0.006; Fig. 2B). A trend, but not statistically significant correlation was also observed with the PL deep layer (r = −0.51,

p = 0.09; Fig. 2B). There was no correlation between the working memory capacity and densities of varicosities in the IL superficial and deep layers (Fig. 2 C). We then investigated alterations of TH expression in the PFC of rats exposed to chronic stress. Reductions of TH expression in chronically stressed rats (n = 10) compared to that in stress-naive rats (n = 12) were observed in the PL superficial and deep layers (stress, F1,80 = 21.7, p < 0.001; subregions, F3,80 = 2.55, p = 0.061; stress × subregions, F3,80 = 3.84, p = 0.013; post hoc test, p = 0.001 and p = 0.007 in the superficial and deep layers, respectively; Fig. 2D). In contrast, reductions of TH expression were small and not statistically significant in the IL superficial and deep layers (Fig. 2D). Among 10 rats exposed to chronic stress, TH expression in the PL superficial layer of a half of them (n = 5; denoted “Low TH group” hereafter; Fig. 2E) was below the lowest level of TH expression in stress-naive rats, whereas TH expression in the other half of the chronically stressed rats (n = 5; denoted “High TH group” hereafter; Fig. 2E) was still within the level comparable to that of stress-naive rats. Distinct patterns of working memory performance and its modulation by chronic stress in the SDA task were emerged between the Low and High TH Groups (Fig. 2F–H). In the High TH Group, chronic stress worsened performance at the 30 s delay (stress, F1,12 = 5.77, p = 0.024; delay, F2,12 = 16.5, p < 0.001;

Y.-A. Lee, Y. Goto / Behavioural Brain Research 286 (2015) 122–127

125

Fig. 2. Chronic stress modulation of TH expression in the PFC and its correlation with working memory function. (A) Representative examples of TH immunohistochemistry in the medial PFC of rats. The squares in the photograph indicate the areas measured as the superficial and deep layers of the PL and IL. Photographs with lower (left photograph, scale bar = 500 ␮m) and higher (right photographs, scale bar = 100 ␮m) magnifications are shown. After subtraction of the background, densities of varicosities were quantified. (B, C) Graphs showing correlations between performance of the SDA task (expressed as difference of percentages of correct responses between the 0 and 30 s delays) and densities of TH varicosities in the PL (B) and IL (C) of stress-naive rats. The dashed and solid lines indicate regression analyses for black (superficial layer) and gray (deep layer) circles, respectively. r: Pearson’s liner correlation coefficient. (D) Quantification of TH varicosities in the whole medial PFC (PL + IL) and those separately analyzed for the superficial (II-III) and deep (V–VI) layers of the PL and IL in chronically stressed and stress-naive rats. *p < 0.001, **p = 0.001, ***p = 0.007. (E) Histograms showing distribution of densities of TH varicosities in the PL superficial layer of chronically stressed (black bars at the bottom) and stress-naive (white bars at the top) rats. The dashed line indicates the point that TH expression is the lowest in stress-naive rats. (F-H) Graphs showing performance of the SDA task of two (Low TH and High TH) groups of rats based on TH expression with chronic stress shown at the bottom of Fig. 2E. Performance before and after chronic stress in the High TH Group (F; *p = 0.001, vs. pre-stress at the 30 s delay) and the Low TH Group (G **p = 0.003, vs. pre-stress at the 0 s delay; ***p = 0.014, vs. pre-stress at the 30 s delay) as well as performance of the SDA task of both groups before stress exposure (H).

stress × delay, F2,12 = 3.43, p = 0.049; post hoc test, p = 0.001; Fig. 2F). In the Low TH Group, chronic stress impaired performance at the 0 s delay, but improved performance at the 30 s delay (stress, F1,12 = 3.94, p = 0.059; delay, F2,12 = 0.36, p = 0.704; stress × delay, F2,12 = 13.3, p < 0.001; post hoc test, p = 0.003 and p = 0.014 at the 0 and 30 s delays, respectively; Fig. 2G). In addition, performance in the High TH Group tested before stress exposure was not significantly different between the 0 and 30 s delays, whereas the Low TH Group tested before stress exposure exhibited delay-dependent decline of performance from the 0 to 30 s delays (unpaired t-test, p = 0.015; Fig. 2H). These results suggest that TH expression in the PL superficial layer may be associated with working memory function, and produce two distinct population of rats in terms of working memory capacity. Moreover, chronic stress may affect working memory by decreasing TH expression in the PL. Chronic stress has been shown to disturb working memory [6]. A deficit of working memory is usually recognized as delaydependent worsening of performance in a working memory test [6,17,18]. In contrast, in this study, we observed that chronic stress

worsened performance in shorter, but not longer delay conditions in the working memory test. Thus, the chronic stress-induced alterations appeared as if brain dysfunction was something other than working memory. Nevertheless, further analysis revealed that this observation was in fact due to existence of heterogeneous population in our tested animals. Such heterogeneity in terms of working memory function of rats was unclear in the stress-naive condition, but became apparent after chronic stress exposure. Our study further revealed that the heterogeneity of working memory function and its modulation by chronic stress was associated with TH expression in the PFC. Genetic regulation of PFC function is still largely unknown. However, for instance, a single nucleotide polymorphism on the COMT gene has been recently shown to impact on cognitive function including working memory by altering DA transmission in the PFC [19]. Murru and colleagues have reported that Sprague-Dawley rats have several polymorphisms on the TH gene [20]. Given that TH genetic backgrounds of Sprague-Dawley rats that Murru have reported (inbred in Italy) and we used in this study (purchased from Charles River in Canada) are different, it has remained unknown that similar polymorphisms

126

Y.-A. Lee, Y. Goto / Behavioural Brain Research 286 (2015) 122–127

exist on the TH gene of rats used in this study. However, it is possible that one of the polymorphisms on the TH gene may regulate TH expression in the PFC, which in turn affects working memory. Several studies investigated the chronic stress effects on PFC catecholamine transmission. The study by Gresch and colleagues have reported decreased tonic, baseline DA tone in the PFC [7]. They also reported that evoked DA release was augmented. This was, however, not truly augmentation, but it appeared as augmentation due to the lower baseline tonic DA tone. Another study by Mizoguchi and colleagues has shown that both tonic DA tone and evoked DA release are decreased [6]. Moreover, Mangiavacchi and colleagues have also reported that 1 week of foot-shock stress exposure attenuates cocaine-induced DA release, whereas stress exposure for 3 weeks attenuates both baseline DA tone and cocaine-induced DA release in the PFC [21]. In accordance with these studies, both chronic stress [13] and 6-hydroxydopamine depletion of DA innervations into the PL [22] induce dendritic spine atrophy in pyramidal neurons, which indirectly suggests that chronic stress decreases DA release. Since TH is a rate-limiting enzyme for synthesis of DA and NE, there is little doubt that DA and NE availabilities are decreased by reduction of TH expression. If DA and NE availabilities are decreased, it is likely that DA and NE release, regardless of basal tonic tone or evoked release, are expected to be decreased. The inhibitor of TH, alpha-methyl-p-tyrosine, has been shown to affect not only synthesis but also release of DA [23]. Thus, the finding of decreased TH expression in the PFC by chronic stress in our study is consistent with the previous studies by Mizoguchi et al. [6] and Mangiavacchi et al. [21]. In contrast, the finding of chronic stress-induced decrease of tonic DA tone but intact evoked DA release by reported by Gresch et al. [7] suggests that chronic stress-induced alterations may also involve DA release regulation mechanisms. Indeed, in this study, we did not examine DA and NE release, such that it has remained unclear and needs to be investigated in a future study whether and in what extent decreased TH expression by chronic stress is translated into decreased DA and NE release. The High TH group is supposed to exhibit less severe stressinduced reduction of DA and NA availabilities than the Low TH group. However, the High TH group with stress exposure showed worse performance in the SDA task than Low TH group. In the SDA test, we observed that one group of rats exhibited high working memory function, such that performance of these rats in the SDA task between the 0 and 30 s delay conditions was comparable. These rats exhibited worsening of performance at the 30 s delay, but not the 0 s delay, after chronic stress exposure, indicating reduction of working memory capacity. The other group of rats exhibited delay-dependent performance in the stress-naive condition. Chronic stress caused an impairment of performance of the SDA task at the 0 s delay, whereas performance was improved in the 30 s delay, in these rats. One explanation for better performance in longer delay than shorter delay conditions with chronic stress in these rats is that they may use different response strategies between before and after chronic stress. The study by Lee and Kesner has shown that the PFC and hippocampus process working memory in parallel, but the PFC primarily retains information for a short duration, whereas the hippocampus involves holding of a much longer duration [24]. Thus, rats with a low working memory capacity may primarily utilize the PFC in a stress-naive condition; however, they may start utilizing the hippocampus to perform the SDA task after chronic stress exposure due to a severe impairment in PFC-dependent working memory function. Collectively, High TH group of rats would exhibit a higher working memory capacity than that of Low TH group in a stress-naive condition. This is supported by a similar, if not identical, pattern of SDA task performance between the Group 1 vs. Group 2 (Fig. 1E/F) and the High vs. Low TH (Fig. 2F/G) rats. Thus, after chronic stress exposure, the

High TH group of rats may still rely on the moderate (or low functioning) PFC-dependent working memory for performance of the SDA task, whereas the Low TH group of rats may switch from PFCdependent to possibly hippocampus-dependent working memory process, especially as consequence of attenuated inhibitory influence of the PFC over limbic structure including the hippocampus [25]. In summary, our study has shown that an association exists between TH expression in the PFC and working memory, and such TH expression in the PFC produces two distinct population of rats in terms of working memory capacity and response to chronic stress in its function. Acknowledgements We thank P. Poirier, H. Yoon, A. Maeda and C. Kalliny for technical assistance. This work was supported by McGill University Health Centre. References [1] Fuster JM. Frontal lobe and cognitive development. J Neurocytol 2002;31:373–85. [2] Arnsten AF. Catecholamine regulation of the prefrontal cortex. J Psychopharmacol 1997;11:151–62. [3] Grace AA, Gerfen CR, Aston-Jones G. Catecholamines in the central nervous system overview. Adv Pharmacol 1998;42:655–70. [4] Fernstrom JD, Fernstrom MH. Tyrosine, phenylalanine, and catecholamine synthesis and function in the brain. J Nutr 2007;137:1539S–47S, discussion 48S. [5] Arnsten AF. Stress signalling pathways that impair prefrontal cortex structure and function. Nat Rev Neurosci 2009;10:410–22. [6] Mizoguchi K, Yuzurihara M, Ishige A, Sasaki H, Chui DH, Tabira T. Chronic stress induces impairment of spatial working memory because of prefrontal dopaminergic dysfunction. J Neurosci 2000;20:1568–74. [7] Gresch PJ, Sved AF, Zigmond MJ, Finlay JM. Stress-induced sensitization of dopamine and norepinephrine efflux in medial prefrontal cortex of the rat. J Neurochem 1994;63:575–83. [8] Ortiz J, Fitzgerald LW, Lane S, Terwilliger R, Nestler EJ. Biochemical adaptations in the mesolimbic dopamine system in response to repeated stress. Neuropsychopharmacol 1996;14:443–52. [9] Smith MA, Brady LS, Glowa J, Gold PW, Herkenham M. Effects of stress and adrenalectomy on tyrosine hydroxylase mRNA levels in the locus ceruleus by in situ hybridization. Brain Res 1991;544:26–32. [10] Xu L, Holscher C, Anwyl R, Rowan MJ. Glucocorticoid receptor and protein/RNA synthesis-dependent mechanisms underlie the control of synaptic plasticity by stress. Proc Natl Acad Sci U S A 1998;95:3204–8. [11] Rocher C, Spedding M, Munoz C, Jay TM. Acute stress-induced changes in hippocampal/prefrontal circuits in rats: effects of antidepressants. Cereb Cortex 2004;14:224–9. [12] Liston C, Miller MM, Goldwater DS, Radley JJ, Rocher AB, Hof PR, et al. Stress-induced alterations in prefrontal cortical dendritic morphology predict selective impairments in perceptual attentional set-shifting. J Neurosci 2006;26:7870–4. [13] Radley JJ, Rocher AB, Miller M, Janssen WG, Liston C, Hof PR, et al. Repeated stress induces dendritic spine loss in the rat medial prefrontal cortex. Cereb Cortex 2006;16:313–20. [14] Goldwater DS, Pavlides C, Hunter RG, Bloss EB, Hof PR, McEwen BS, et al. Structural and functional alterations to rat medial prefrontal cortex following chronic restraint stress and recovery. Neurosci 2009;164:798–808. [15] Lee YA, Goto Y. Chronic stress modulation of prefrontal cortical NMDA receptor expression disrupts limbic structure-prefrontal cortex interaction. Eur J Neurosci 2011;34:426–36. [16] Lee YA, Poirier P, Otani S, Goto Y. Dorsal-ventral distinction of chronic stressinduced electrophysiological alterations in the rat medial prefrontal cortex. Neurosci 2011;183:108–20. [17] Zahrt J, Taylor JR, Mathew RG, Arnsten AF. Supranormal stimulation of D1 dopamine receptors in the rodent prefrontal cortex impairs spatial working memory performance. J Neurosci 1997;17:8528–35. [18] Aultman JM, Moghaddam B. Distinct contributions of glutamate and dopamine receptors to temporal aspects of rodent working memory using a clinically relevant task. Psychopharmacol 2001;153:353–64. [19] Egan MF, Goldberg TE, Kolachana BS, Callicott JH, Mazzanti CM, Straub RE, et al. Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia. Proc Natl Acad Sci U S A 2001;98:6917–22. [20] Murru S, Pani L, Poddie D, Gessa G, Cao A, Pirastu M. Rat tyrosine hydroxylase gene polymorphisms. Brain Res Mol Brain Res 1997;45:345–8. [21] Mangiavacchi S, Masi F, Scheggi S, Leggio B, De Montis MG, Gambarana C. Longterm behavioral and neurochemical effects of chronic stress exposure in rats. J Neurochem 2001;79:1113–21.

Y.-A. Lee, Y. Goto / Behavioural Brain Research 286 (2015) 122–127 [22] Wang HD, Deutch AY. Dopamine depletion of the prefrontal cortex induces dendritic spine loss: reversal by atypical antipsychotic drug treatment. Neuropsychopharmacol 2008;33:1276–86. [23] Fairbrother IS, Arbuthnott GW, Kelly JS, Butcher SP. In vivo mechanisms underlying dopamine release from rat nigrostriatal terminals: I. Studies using veratrine and ouabain. J Neurochem 1990;54:1834–43.

127

[24] Lee I, Kesner RP. Time-dependent relationship between the dorsal hippocampus and the prefrontal cortex in spatial memory. J Neurosci 2003;23: 1517–23. [25] Goto Y, Grace AA. Dopamine modulation of hippocampal-prefrontal cortical interaction drives memory-guided behavior. Cereb Cortex 2008;18: 1407–14.