Orexins in the Midline Thalamus Are Involved in the Expression of Conditioned Place Aversion to Morphine Withdrawal

Physiology & Behavior 102 (2011) 42–50

Contents lists available at ScienceDirect

Physiology & Behavior

journal homepage: www.elsevier.com/locate/phb

Orexins in the midline thalamus are involved in the expression of conditioned place aversion to morphine withdrawal

Yonghui Li a, Huiying Wang b, Keke Qi b, Xiaoyu Chen c,SaLic, Nan Sui b, Gilbert J. Kirouac c,d,⁎ a Key Laboratory of Mental Health, Institute of Psychology, Chinese Academy of Sciences, Beijing, China and Faculty of Dentistry, University of Manitoba, Winnipeg, Manitoba, Canada b Behavioral Pharmacology Laboratory, Institute of Psychology, Chinese Academy of Sciences, Beijing, China c Department of Oral Biology, Faculty of Dentistry, University of Manitoba, Winnipeg, Manitoba, Canada d Department of Psychiatry, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada article info abstract

Article history: Previous studies have implicated the bed nucleus of the stria terminalis, central nucleus of the amygdala and Received 19 July 2010 the shell of the nucleus accumbens (collectively called the extended amygdala) as playing an important role in Received in revised form 1 October 2010 mediating the aversive emotion associated with opioid withdrawal. The paraventricular nucleus of the Accepted 8 October 2010 thalamus (PVT) provides a very dense input to the extended amygdala, and the PVT is densely innervated by orexin neurons, which appear to be involved in producing some of the physical and emotional effects Keywords: associated with morphine withdrawal. In the present study, we confirm that the PVT is densely innervated by Orexin orexin fibers, whereas the regions of the extended amygdala associated with the effects of morphine Hypocretin Morphine withdrawal withdrawal are poorly innervated. Microinjections of the orexin-1 receptor (OX1R) antagonist SB334867 or Thalamus the orexin-2 receptor (OX2R) antagonist TCSOX229 at doses of 5.0 or 15.0 μg into the PVT region did not affect Extended amygdala the acquisition of the conditioned place aversion (CPA) nor the physical effects produced by naloxone- Conditioned place aversion precipitated morphine withdrawal. In contrast, microinjections of TCSOX229 (15.0 μg) in the PVT region significantly attenuated the expression of naloxone-induced CPA while microinjections of SB334867 at the same dose had no effect. The results from these experiments indicate a role for OX2R in the PVT on the expression of CPA associated with morphine withdrawal. Orexins may mediate the aversive effects of morphine withdrawal by engaging the extended amygdala indirectly through the action of orexins on the PVT. © 2010 Elsevier Inc. All rights reserved.

1. Introduction Neuropeptides called orexins have received attention for their importance in regulating behaviors related to drug reward [3,4]. Several Opioid drugs like heroin and morphine produce euphoric effects studies have also shown that orexins may be involved in the withdrawal that can lead to chronic drug use and dependence. In contrast, effects of morphine. First, morphine withdrawal was found to increase abstinence from opiates in dependent individuals produces negative the expression of c-Fos gene or protein in orexin neurons as well as to physiological (nausea, diarrhea, vomiting, tachycardia and sweating) increase orexin mRNA level in the hypothalamus [5–7].Second,orexin and emotional (irritability, anxiety, and dysphoria) states that are knock-out mice display fewer jumps, tremor and diarrhea in naloxone- extremely unpleasant to the drug user [1]. While the physiological precipitated morphine withdrawal [7]. Third, systemic administration of effects produced by opiate withdrawal dissipate over several days, the the orexin-1 receptor (OX1R) antagonist SB334867 attenuated beha- unpleasant emotional effects produced by opioids remain for weeks viors associated with morphine withdrawal in mice, an effect that or months and represent a major factor contributing to relapse [1,2]. appeared to be mediated by a decrease in the activity of neurons in the Consequently, there has been significant interest in identifying the shell of the nucleus accumbens [6]. The same paper reported that only a brain mechanisms that contribute to the negative emotional states few orexin neurons in the hypothalamus innervate the nucleus produced by abstinence of opioid and other addictive drugs with the accumbens, suggesting that orexins influence neurotransmission in hope that novel pharmacological targets can be identified to prevent this area of the striatum via an indirect projection [6]. relapse. The paraventricular nucleus of the midline thalamus (PVT) is densely innervated by orexin fibers [8,9] and is a major source of glutamatergic afferents to the shell of the nucleus accumbens, dorsolateral bed nucleus of the stria terminalis and the central nucleus of the amygdala [10–13]. ⁎ Corresponding author. Tel.: +1 204 977 5696; fax: +1 204 789 3913. These areas of the basal forebrain are collectively called the extended E-mail address: [email protected] (G.J. Kirouac). amygdala [14–17] and represent an anatomical macrostructure important

0031-9384/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2010.10.006 Y. Li et al. / Physiology & Behavior 102 (2011) 42–50 43 for mediating both the physical and emotional effects of morphine containing 5% normal donkey serum, 0.3% Triton X-100 and 0.1% sodium withdrawal [6,18–21]. In fact, the entire extended amygdala is a large azide followed by incubation in rabbit antiserum against OXA (1:3000, and complex area of the basal forebrain involved in the integration of diluted in pre-incubation solution; Chemicon, Temacula, CA; catalogue the motivational and physiological responses associated with emotions #AB3704) or goat antiserum against OXB overnight (1:500, diluted in [22–24]. Subregions of the extended amygdala including the shell of the pre-incubation solution; Santa Cruz Biotech Inc., Santa Cruz, CA; nucleus accumbens, lateral bed nucleus of the stria terminalis and the catalogue No. sc 8071). The sections were then rinsed for 3×10 min central nucleus of the amygdala have been identified as being especially in PBS, followed by placing the sections in either biotinylated donkey important for mediating the different effects of morphine withdrawal anti-rabbit or donkey anti-goat (1:500; Jackson Immunoresearch, West [6,18,19,21,25,26]. However, it is questionable if the release of orexins in Grove, PA) antisera for 2 h. Then sections were rinsed thoroughly and the extended amygdala mediates the effect of morphine withdrawal incubated with an avidin-biotin complex (ABC) solution (Elite ABC kit; because most of the critical areas of the extended amygdala activated by Vector Laboratories, Burlingame, CA). After three more rinsing steps, morphine withdrawal appear to be poorly innervated by orexin fibers sections were reacted with 3,3 diaminobenzidine tetrachloride (DAB) [6,27]. As such, the PVT, with its dense projections to those areas of the with nickel intensification (Vector DAB Kit, Vector Laboratories extended amygdala activated by morphine withdrawal, represents a Burlingame, CA) to produce a black reaction product. The DAB reaction potential site where orexins could act to produce the physiological and was terminated by rinsing the sections with PBS and the sections were emotional effects associated with morphine withdrawal. mounted on slides, air dried and cover-slipped with Fluka DPX The present study was done to determine if blockade of orexin mountant. Sections of the forebrain were examined and photographed receptors in the PVT with antagonists for the OX1R or the orexin-2 under magnification using an Olympus BX51 microscope equipped with receptors (OX2R) attenuated the symptoms associated with morphine a digital camera (SPOT RT Slicer, Diagnostic Instruments Inc, Sterling withdrawal. This was done by examining the effects of orexin receptor Heights, MI). Calibration bars were inserted using SPOT software antagonists on the acute physical symptoms and conditioned place (Version 3.2, Diagnostic Instruments) and images were transferred to aversion (CPA) produced by naloxone-induced precipitated morphine Adobe Photoshop 5.5 to optimize light and contrast levels. withdrawal. The CPA paradigm involves Pavlovian conditioning where the negative emotional state an animal experiences during morphine 2.3. Behavioral experiments withdrawal is paired with the particular environment where the emotional state occurs [28]. Re-exposure of the animal to the paired 2.3.1. Surgery environment leads to the experience of a negative emotional state and Rats were anesthetized with equithesin (0.3 ml/100 g, i.p.) and the avoidance of that environment [28]. In this case, CPA can be used to placed in a Stoelting stereotaxic frame. Stainless steel guide cannula investigate the neural mechanisms involved in the expression of (23 gauge, Plastics One, Roanoke, VA, USA) were unilaterally negative emotions link to opioid withdrawal. We also compared the implanted into the posterior aspect of the PVT (3.1 mm posterior to density of orexin-A (OXA) and orexin-B (OXB) fibers in the PVT and the bregma, 1.3 mm lateral to the midline, and 4.0 mm ventral to the extended amgydala to re-examine the extent of orexin innervation of skull, at 10˚ angle, with the incisor bar set at 3.3 mm below intraaural key areas of the forebrain linked to the negative emotional and aversive line). The guide cannula was secured with three small screws attached state associated with morphine withdrawal. to the skull and embedded in dental cement and a capped stylet was inserted. All rats were treated with penicillin (80,000 units) to 2. Materials and methods prevent infection. The rats were allowed to recover for 7–10 days and were handled every other day to reduce stress associated with 2.1. Animals and housing handling.

Male Sprague–Dawley rats (240–260 g, n=44) were housed 2.3.2. Drugs and microinjections individually in translucent plastic cages (37 cm×24 cm×19 cm) in a Morphine hydrochloride (Qinghai Pharmaceutical, China), naloxone colony room on a 12 h/12 h light dark cycle (light on at 07:00) with hydrochloride (Sigma, USA) and the OX2R antagonist TCSOX229 (Tocris, controlled temperature (22–24 °C) and humidity (40–60%). The rats UK) were dissolved in sterile physiological saline (0.9% NaCl), whereas were allowed free access to food and water in the home cage. All rats the OX1R antagonist SB334867 (Tocris) was dissolved in 100% DMSO. were gently handled three times to adapt them to the stress of being Previous studies using an avoidance memory task have shown that handled, and behavioral manipulations were conducted in the light DMSO does not alter the antagonist properties of SB [29–31] nor that it phase (08:00–18:00). The experiments were done following the produces behavioral effect when compared to saline [32].Morphineand guidelines provided by the National Institutes of Health Guide for Care naloxone were administered subcutaneously in volumes of 1.0 ml/kg and Use of Laboratory Animals (1986), Regulation for the Adminis- body weight. TCSOX229 and SB334867 were microinjected into the PVT tration of Affairs Concerning Experimental Animals (China) and the (0.5 μl) through an injector cannula (28 gauge) which protrudes 2.0 mm Canadian Council on Animal Care. The experimental protocols were below the guide cannula. Infusions were delivered with a 5.0 μl Hamilton approved by the local ethics committees at the Institute of Psychology, microsyringe attached to the injector cannula with polyethylene tubing. Chinese Academy of Sciences and the University of Manitoba. The injection was done at the rate of 0.25 μl/minover2minwiththe injector cannula remaining in the guide cannula for another minute to 2.2. Immunohistochemical experiments prevent backflow. The stylet was then placed back into the guide cannula and the rat was returned to its home cage. Four naïve rats were deeply anesthetized with the mixture of ketamine (100 mg/kg, i.p.) and xylazine (5 mg/kg, i.p.). The animals 2.3.3. Behavioral testing and experimental design were perfused transcardially with 150 ml 0.9% saline solution with 0.6% The place conditioning apparatus consisted of a rectangular two- heparin followed by 400 ml ice-cold 4% paraformaldehyde in 0.1 M compartment plastic chamber 80 (length)×40 (width)×40 cm phosphate buffer (PB, PH 7.4). Brains were removed and post-fixed in (height) separated with a guillotine door. One compartment had the same solution for 1 h and then cryoprotected in 20% sucrose walls with red stripes and a smooth surfaced floor, whereas the other phosphate buffered saline (PBS, PH 7.4) at 4 °C overnight. Sections of the compartment had black walls and a grid floor. The apparatus was brain were cut at 50 μm using a cryostat and collected in PBS. placed in a room with dim light provided by three incandescent bulbs Immunohistochemical reactions were done on free-floating sections (15 W). The location and movement of rats were monitored by a at room temperature. Brain sections were pre-incubated in 0.1 M PBS video camera suspended from the ceiling and analyzed for time and 44 Y. Li et al. / Physiology & Behavior 102 (2011) 42–50 distance in each compartment using tracking Software (Taiji Software orexin receptors in the PVT prevented the expression of CPA. The Company, Beijing). assignment of rats to their respective groups was counterbalanced to As shown in a diagram of the experimental protocol (Fig. 1), the prevent same group of animals from receiving the same drug twice. same rats served as subjects for both the CPA acquisition and expression experiments. Rats received twice a day injections of morphine for 4 days with escalating daily doses (10.0, 20.0, 30.0 and 2.3.4. Cannula placement verification 40.0 mg/kg, s.c.). On day 1, rats received their two daily administra- At the end of the experiment, all rats were deeply anesthetized tions of morphine without being exposed to the conditioning with chloral hydrate (40 mg/kg), perfused transcardially with chamber. On day 2 of morphine treatment, rats were adapted to the heparinized saline followed by 4.0% paraformaldehyde in 0.1 M PB. conditioning apparatus by placing them in the chamber for 30 min to Coronal sections from the injection site were obtained and the freely explore the new environment. On day 3, the compartment locations of the cannula tips in relation to the PVT were assessed on preference was assessed by placing the rats in the central part of sections stained for OXA using immunohistochemistry as previously chamber for 15 min and recording the time spent in each compart- done in our laboratory [8]. ment (referred as the preconditioning). On days 2 and 3 the rats were placed in the chamber at least 4 h after the morphine injection. Day 4 comprised the conditioning (acquisition) phase of the experiments 2.3.5. Statistical analysis where morphine treatment was followed by naloxone in rats that All data are expressed as mean±SEM and analyzed using SPSS 13.0 had received a microinjection of the orexin antagonist or vehicle software for Windows. The data on weight loss, defecation and in the PVT region. Rats received a microinjection of either the locomotor activity were analyzed by one-way ANOVA, and the OXR1 antagonist SB334867 (5.0 or 15.0 μg, n=8), OXR2 antagonist acquisition and expression of CPA were analyzed by two-way ANOVA TCSOX229 (5.0 or 15.0 μg, n=8) or vehicle (n=8) in the PVT region with “test session” (preconditioning, post-conditioning, expression) as a 30 min before receiving a naloxone injection. A group of morphine- within-subject factor and “drug treatment” (saline; SB334867 and treated rats (n=8) was injected with saline (s.c.) instead of naloxone TCSOX229 at 5 and15 μg) as a between-subject factor. When the ANOVA and vehicle in the PVT region as a control group to assess the extent of was found significant, Dunnett post-hoc tests were applied to determine the naloxone-induced morphine withdrawal. During the conditioning which groups were statistically different. Percentage of wet-dog shake training session, an unbiased design was used in which half of the rats and jump were analyzed by Chi Square analysis. in each group received naloxone (0.5 mg/kg, s.c., 2 h after the last morphine injection) and was placed in the preferred compartment while the other half received naloxone paired with the non-preferred 3. Results compartment. The conditioning session lasted 30 min and weight loss, number of defecation, number of jumps, and number of wet-dog 3.1. Orexin fibers in the extended amygdala shakes were quantified to assess the acute effect of morphine withdrawal [33]. The withdrawal related behaviors were quantified Fig. 2 shows the distribution of OXA and OXB fibers in the different from video recordings by two experimenters blind to group identity of regions of the extended amygdala and the PVT. Qualitatively, the the rats. Because of the high variability of jump and wet-dog shake distribution and density of OXA and OXB immunostained fibers in the behaviors among the animals, percentage of the rats with jump or forebrain were found to be the same. A weak density of orexin fibers wet-dog shake was calculated to assess the effect of the drug. On day was observed in the ventral part of the nucleus accumbens shell and 5, a post-conditioning test session was done by placing the drug-free the lateral subnucleus of the central nucleus of the amygdala (area is rat in the chamber with free access to both compartments for 15 min largely devoid of fibers or contains scattered fibers with only a few to identify the subjects that still expressed CPA (referred as the post- enlargements or buttons). A moderate density of orexin fibers was conditioning). found in the dorsolateral region of the shell of the nucleus accumbens, After the post-conditioning test, the data showed that all the medial subdivision of the bed nucleus of the stria terminalis and the groups of animals developed significant CPA, indicating both OX1R medial subnucleus of the central nucleus of the amygdala (larger and OX2R antagonists had no effect on the acquisition of CPA. The rats areas of neural tissue lacking fibers and having fibers with enlarge- were retested on day 6 to make sure that the CPA was established and ments or terminals). The dorsolateral subdivision of the bed nucleus only rats showing a robust CPA (the time difference between pretest of the stria terminalis and the lateral subnucleus of the central nucleus and retest was more than 100 s) were used to examine the effects of of the amygdala, which contain a high density of dynorphin and orexin receptor antagonists on the expression of CPA. Consequently, corticotropin releasing factor (CRF) neurons [34], were nearly devoid only rats that showed a stable CPA expression were used for the of orexin fibers except for occasional fibers seen crossing these expression experiments. The rats were then re-assigned into groups subnuclei. In contrast, as we previously reported [8,9], orexin fibers that received microinjection of either SB334867 (15 μg, n=8), were found to be very densely concentrated in all regions of the PVT TCSOX229 (15 μg, n=8) or vehicle (n=8) into the PVT region (most of the area covered with fibers containing many enlargements 30 min before placing the animals in the chamber to see if blocking of and ending in clusters of terminals).

Fig. 1. Diagram showing the sequence of treatments and tests used for the morphine withdrawal CPA acquisition and expression experiments. D1–D7 represent the days associated with each procedure. Y. Li et al. / Physiology & Behavior 102 (2011) 42–50 45

Fig. 2. Images showing the distribution of orexin-A (OXA, left side) and orexin-B (OXB, right side) fibers in the nucleus accumbens (A,B), bed nucleus of the stria terminalis (C,D), central nucleus of the amygdala (E,F) and paraventricular nucleus of the thalamus (G,H). Note that the boundary between the paraventricular nucleus of the thalamus and other nuclei is defined by the high density of OXA (G) and OXB (H) fibers as previously shown [8]. ac, anterior commissure; CeC, capsular subnucleus of the central nucleus of the amygdala; CeL, lateral subnucleus of the central nucleus of the amygdala; CeM, medial subnucleus of the central nucleus of the amygdala; Hb, habenular complex; NaC, core of the nucleus accumbens; NaS, shell of the nucleus accumbens; STLD; dorsal lateral subregion of the bed nucleus of the stria terminalis; STMA, anterior medial subregion of the bed nucleus of the stria terminalis; STLP, lateral posterior subregion of the bed nucleus of the stria terminalis; 3V, third ventricle. 46 Y. Li et al. / Physiology & Behavior 102 (2011) 42–50

3.2. Cannula placement there was no treatment effect (saline, SB334867 and TCSOX229 treated rats) in rats receiving naloxone on weight loss (Fig. 4D, F(4,35)=0.347, Cannula placements within 0.5 mm of the PVT were considered pN0.05), locomotor activity (Fig. 4E, F(4,35)=0.236, pN0.05), jumping acceptable [35] and no subjects were removed from the data analysis (Fig. 4A,χ2 =2.96,df=4,pN0.05) and wet-dog shaking (Fig. 4B,χ2 =4.0, because of inaccurate cannula placements. The tips of the injector df=4, pN0.05). cannulae either localized in the posterior PVT or at the boundary region of the PVT with the mediodorsal nucleus and the intermedio- 3.4. Effect of orexin receptor antagonists on the development of CPA dorsal nucleus of the thalamus. Examples of some of the placements are shown in Fig. 3. As discussed in our previous papers [35], these As shown in Fig. 5, morphine-treated rats that received naloxone placements were chosen to limit the damage to the PVT caused by the developed CPA compared to the ones that received saline as shown by insertion and removal of the injector cannula. a significant interaction effect (Fig. 5A, F(1,15)=4.74, pb0.05). The effect of microinjections of orexin receptor antagonists in the PVT 3.3. Effect of orexin receptor antagonists on the withdrawal related region on the development of CPA is shown in Fig. 5B. The ANOVA behaviors revealed that there was no main effect of the “drug treatment” (Fig. 5B, F(4,35)=0.09, pN0.05), nor significant interaction effect of As depicted in Fig. 3, morphine-treated rats which received naloxone “drug treatment” and “test session” (Fig. 5B, F(4,35)=0.33, pN0.05) produced more defecation (Fig. 4A, F(1,15)=112.44, pb0.0001), more on the development of CPA, but there was a main effect of “test weight loss (Fig. 4B, F(1,15)=73.45, pb0.0001) and less locomotor session” (Fig. 5B, F(1,35)=31.74, pb0.01). This suggests that all the activity (Fig. 4C, F(1,15)=19.99, pb0.0001) compared to those which groups acquired significant CPA and that the orexin receptor received saline injections. In addition, the frequency of jumping antagonists had no effect on the acquisition of CPA. (Fig. 4D,χ2 =16.16, df=5, pb0.05) and wet-dog shaking episodes (Fig. 4E,χ2 =8.62,df=5, pb0.05) was greater in naloxone treated rats. 3.5. Effect of orexin receptor antagonists on the expression of CPA The expression of these behaviors indicates that naloxone induced a robust withdrawal effect in the morphine-treated rats. Two doses of the The effect of orexin receptor antagonists on the expression of CPA is OX1R or OX2R antagonists were injected into PVT to examine the effect presented in Fig. 6. The two-way ANOVA showed that there is a significant oftheorexinreceptorblockadeinthePVTonthephysicalwithdrawal “test session” by “drug treatment” interaction effect on the time spent in symptoms. The one-way ANOVA revealed that there is a significant main the naloxone-paired compartment (F(4,42)=2.73, pb0.05). Simple effect of “drug treatment” on defecation among groups in rats treated effect analysis revealed that there were significant differences with SB334867 and TCSOX229 (Fig. 4A, F(4,35)=2.71, pb0.05). The among the three test sessions in the vehicle (F(2,42)=15.13, pb0.01), post-hoc analysis revealed that the high dose (15.0 μg) of the OX2R SB (F(2,42)=13.69, pb0.05) and TCS groups (F(2,42)=13.11, pb0.01). antagonist TCSOX229 produced a small but significant decrease in Further analysis showed that the rats receiving vehicle (pb0.01) and SB defecation compared to the vehicle treated rats (pb0.05). However, (pb0.01) spent less time on the naloxone-paired side during the expression test session than that during the preconditioning session. In contrast, there is no significant difference of the time spent on the naloxone-paired side between the preconditioning and expression sessions in the TCS-treated rats which indicates that TCS impaired the expression of CPA. Further support for this conclusion is provide by another simple effect analysis showing significant difference among “drug treatment” groups during the expression test session (F(2,21)= 2.56, pb0.05) in addition to the post-hoc analysis which reveals that the rats in TCSOX229 group spent more time on the naloxone-paired side compared to the ones in the vehicle group (pb0.05).

4. Discussion

The CPA paradigm involves both aversive memory formation (acquisition of CPA) and aversive memory retrieval (expression of CPA). In this study, blocking of neither the OX1R nor OX2R in the PVT region had effect on the acquisition of CPA produced by naloxone- induced morphine withdrawal, suggesting that the release of orexins in the PVT is not important for learning the relationship between the aversive state associated with morphine withdrawal and the context where this state occurred. Furthermore, blocking of orexin receptors in the PVT region had no effect on the physical symptoms associated with acute morphine withdrawal. Based on these results, we conclude that the release of orexins in the PVT during morphine withdrawal is not involved in the production of the negative emotional and physiological states associated with acute morphine withdrawal. In contrast, blocking of the OX2R in the PVT region, but not OX1R, attenuated the expression of CPA which indicates that the release of orexins in the PVT is important in the expression of CPA related with re-exposure to contextual cues associated with morphine withdrawal. These results are consistent with mounting evidence pointing to a role Fig. 3. Images of two examples of injector cannula placements in or immediately for the PVT in conditioned behavioral responses [36–42]. adjacent to the paraventricular nucleus of the thalamus (PVT) in sections stained for orexin-A. As we previously reported, orexin-A ﬁbers clearly show the boundaries of the The PVT, mediodorsal, intermediodorsal, centromedial, paracen- PVT within the midline thalamus [8]. Scale bar 200 μm. tral nuclei, as well as the habenula, are all within the general region Y. Li et al. / Physiology & Behavior 102 (2011) 42–50 47

Fig. 4. Effect of microinjections of vehicle (Veh), SB334867 at 5.0 or 15.0 μg (SB5 and SB15) or TCSOX229 at 5.0 or 15.0 μg (TCS5 and TCS15) in the paraventricular nucleus of the thalamus on the number of defecation (A), weight loss (B), locomotor activity (C), jumping (D), and shaking (E) produced by naloxone (Nal) induced morphine withdrawal. #pb0.0001, compared to saline (Sal/Veh); *pb0.05, compared to Nal/Veh. where the orexin antagonists were injected in the midline thalamus. compared to other thalamic nuclei, the PVT contains a high density However, several facts suggest that the behavioral effects of the OX2R of orexin ﬁbers, which implies that this nucleus is one of the most antagonist were largely mediated by actions on the PVT. First, important targets for this peptide [8]. Second, the PVT contains a

Fig. 5. Development of conditioned place aversion (CPA) produced by naloxone-induced morphine withdrawal (A) and the effect of vehicle (Veh), SB334867 at 5.0 or 15.0 μg (SB5 and SB15) or TCSOX229 at 5.0 or 15.0 μg (TCS5 and TCS15) administration in the paraventricular nucleus of the thalamus (PVT) on the acquisition of CPA (B). Note that microinjections of orexin antagonists in the PVT did not prevent the development of CPA (B). *pb0.05, compared to preconditioning. 48 Y. Li et al. / Physiology & Behavior 102 (2011) 42–50

of the OX1R and OX2R antagonists in the PVT region did not attenuate the physical effects of morphine withdrawal or the acquisition of CPA. This indicates that the release of orexins in the PVT is not important for producing the physical and emotional effects experienced during acute morphine withdrawal. In contrast to studies examining the neural mechanisms associated with the acquisition of CPA to morphine withdrawal, little is known about the mechanisms associated with the expression of CPA. For example, c-Fos expression was not increased in the extended amygdala of animals re-exposed to the context of morphine withdrawal [21,26]. One possible explanation for these findings is that exposure of an animal to the context where morphine withdrawal occurred does not strongly activate neurons in the extended amygdala to a sufficient extent to produce detectable changes in c-Fos expression in these neurons. The fact that expression of c-Fos in the brain is much more intense following Fig. 6. The effect of vehicle (Veh), SB334867 (SB) or TCSOX229 (TCS) administration naloxone-precipitated morphine withdrawal than to the re-exposure of (15 μg) in the paraventricular nucleus of the thalamus on the expression of conditioned place aversion (CPA). *pb0.05, compared to preconditioning session, #p b0.05 compared an animal to a context associated with the withdrawal is consistent with to Veh. this idea [26]. As shown recently following lesions of parts of the amygdala[56], it is not unreasonable to expect that the extended amygdala would play a role in the expression of morphine withdrawal moderately high expression of mRNA and protein for the OX2R considering the well-defined role of this area of the forebrain in the whereas nuclei immediately adjacent to the PVT do not express OX2R acquisition of CPA as well as aversive/avoidance behaviors [57,58]. [43,44]. Third, previous work in our laboratory has shown that The PVT, with its strong direct projection to the extended amygdala, is microinjections of orexins in the PVT region produce anxiety-like in a key anatomical position to influence brain regions involved in effects whereas injections made more laterally do not [45] indicating emotions and avoidance [10]. Previous work by our laboratory has shown the orexin receptors mediating the behavioral effects of orexins in the that orexins act at the PVT to regulate negative emotional states thalamus are near the midline. While it is possible that the effects [35,45,59] predominately through an OX2R mediated mechanism [45]. observed in the present study may have been due to the action of the The main finding of the present study is that microinjections of an OX2R OX2R antagonist at other areas of the brain possibly through the brain antagonist in the PVT region attenuated the expression of CPA produced ventricular system, this is unlikely because the OX1R antagonist, by exposure of rats to the context of precipitated morphine withdrawal. which was delivered in the same manner as the OX2R antagonist, did While the signals for both OX1R and OX2R mRNA have been found to be not block the acute somatic effects of morphine withdrawal as was relatively high in the PVT [43,60],theproteinexpressionforOX2R reported previously when the same OX1R antagonist was given appears to be much greater than that for OX1R [44,61]. Consequently, our systemically [6]. Although it is not possible to completely exclude a results showing that microinjections of the OX1R antagonist in the PVT contribution by other midline nuclei in the present study, the PVT region did not attenuate the expression of CPA in the present study are appears to be the most likely candidate for mediating the behavioral not entirely surprising considering that the PVT contains a low expression effects of OX2R antagonist observed here. of OX1R [61] and that orexins produce potent excitatory effects on most Experimental evidence shows that orexins are important for neurons in the PVT by actions involving the OX2R [62–64].The regulating behaviors related to drug reward and goal directed behaviors availability of a specific antagonist for the OX1R has led to the [3,46,47] and as reviewed elsewhere [4,48]. However, several lines of accumulation of evidence pointing to the use of agents that block the evidence indicate that orexin neurons may also be involved in the OX1R as a potentially useful approach to prevent relapse to a number of aversive effects of morphine withdrawal [5–7]. One recent study addictive substances (see reviews by [48,65]). In contrast, much less is showed that blockade of OX1R in the locus coeruleus attenuated the known concerning the role of the OX2R as it relates to brain function as somatic signs of morphine withdrawal [49]. Another study has well as the mechanisms associated with drug addiction. The results from implicated orexins and the shell of the nucleus accumbens as being this study suggest that the OX2R may also be an important pharmaco- important for mediating the acute withdrawal symptoms of morphine logical target for drug addiction especially as it relates to the negative [6]. In addition to the shell of the nucleus accumbens, other areas of the emotional state associated with abstinence. extended amygdala have been linked to the somatic signs and aversive Experimental evidence shows that PVT neurons are activated by effects associated with morphine withdrawal [18,19,50–55]. However, exposure to contextual cues associate with positive and negative it is questionable whether the withdrawal effects of morphine are emotional states [36–38] including response to exposure to the same mediated by a direct projection from orexin neurons to the extended context associated with psychostimulants and ethanol administrations amygdala. For example, a light orexin fiber density was reported for the [39–41]. Similar to the present study, the functional importance of the dorsomedial part of the nucleus accumbens whereas the rest of the PVT in context-related behavioral responses is also indicated by a recent nucleus accumbens is not innervated [27].Wealsoconfirm the paucity paper showing that lesions of the PVT prevented the context-induced of orexin fibers in the lateral subnucleus of the central nucleus of the reinstatement of ethanol seeking even if similar lesions had no effect on amygdala and the dorsolateral subnucleus of the bed nucleus of the stria the acquisition of this behavior [42]. As such, we interpret the results of terminalis as previously reported [27]. As shown in the present paper as the present study as an indication that the release of orexins in the PVT well as previous studies [6,27], the anatomical connections between plays a role in producing the avoidance of the context in which morphine orexin neurons and regions of the extended amygdala associated with withdrawal occurred. It is reasonable to propose that exposure to the the acute withdrawal effects of morphine are weak and are unlikely to context where morphine withdrawal occurred leads to the release of be involved in mediating the withdrawal effects of morphine. The PVT, orexins in the PVT and possibly the activation of PVT glutamatergic with its dense orexin innervation [8,9] and unique connections to the neurons [62–64] that innervate regions of the extended amygdala which shell of the nucleus accumbens and the central extended amygdala [10], regulate negative emotions and avoidance behavior [10]. The mecha- represents a possible site where orexin release could mediate some of nism as proposed is consistent with previous work in our laboratory the acute withdrawal effects of morphine. However, this does not showing that orexins act at the PVT to enhance negative emotions and appear to be the case since in this paper we report that microinjections avoidance behavior[35,45,59] through mechanisms involving the Y. Li et al. / Physiology & Behavior 102 (2011) 42–50 49 release of CRF and opioid peptides that act at CRF and kappa receptors in [24] Davis M, Walker DL, Miles L, Grillon C. Phasic vs sustained fear in rats and humans: role of the extended amygdala in fear vs anxiety. Neuropsychopharma- the brain [45]. Clearly, more studies will be required to provide cology 2009;35:105–35. unequivocal support for the idea that an orexin-PVT-extended amygdala [25] Harris GC, Aston-Jones G. Enhanced morphine preference following prolonged system is involved in the avoidance of the context associated with the abstinence: association with increased Fos expression in the extended amygdala. Neuropsychopharmacology 2003;28:292–9. aversive state produced by morphine withdrawal. [26] Frenois F, Stinus L, Di Blasi F, Cador M, Le Moine C. A specific limbic circuit underlies opiate withdrawal memories. J Neurosci 2005;25:1366–74. Acknowledgements [27] Baldo BA, Daniel RA, Berridge CW, Kelley AE. Overlapping distributions of orexin/ hypocretin- and dopamine-beta-hydroxylase immunoreactive fibers in rat brain regions mediating arousal, motivation, and stress. J Comp Neurol 2003;464:220–37. The study was supported by the Canadian Institutes of Health [28] Stinus L, Le Moal M, Koob GF. Nucleus accumbens and amygdala are possible Research (CIHR; MOP68909 to G.J.K.); the National Basic Research substrates for the aversive stimulus effects of opiate withdrawal. Neuroscience – Program grants and the Chinese Academy of Sciences grant 1990;37:767 73. [29] Akbari E, Motamedi F, Naghdi N, Noorbakhshnia M. The effect of antagonization of (2009CB522002, KSCXI-YW-R-68 to N.S.); and the NSFC grants and the orexin 1 receptors in CA1 and dentate gyrus regions on memory processing in Project for Young Scientists Fund, Institute of Psychology, CAS (31070911, passive avoidance task. Behav Brain Res 2008;187:172–7. 07CX081008 to Y.L.). [30] Akbari E, Naghdi N, Motamedi F. Functional inactivation of orexin 1 receptors in CA1 region impairs acquisition, consolidation and retrieval in Morris water maze task. Behav Brain Res 2006;173:47–52. References [31] Akbari E, Naghdi N, Motamedi F. The selective orexin 1 receptor antagonist SB- 334867-A impairs acquisition and consolidation but not retrieval of spatial [1] O'Brien CP. Review. Evidence-based treatments of addiction. Philos Trans R Soc memory in Morris water maze. Peptides 2007;28:650–6. Lond B Biol Sci 2008;363:3277–86. [32] Naghdi N, Asadollahi A. Genomic and nongenomic effects of intrahippocampal [2] Koob GF. Dynamics of neuronal circuits in addiction: reward, antireward, and microinjection of testosterone on long-term memory in male adult rats. Behav emotional memory. Pharmacopsychiatry 2009;42(Suppl 1):S32–41. Brain Res 2004;153:1–6. [3] Boutrel B, de Lecea L. Addiction and arousal: the hypocretin connection. Physiol [33] Pinelli A, Trivulzio S. Quantitative evaluation of opioid withdrawal signs in rats Behav 2008;93:947–51. repeatedly treated with morphine and injected with naloxone, in the absence or [4] Aston-Jones G, Smith RJ, Moorman DE, Richardson KA. Role of lateral hypotha- presence of the antiabstinence agent clonidine. J Pharmacol Toxicol Meth 1997;38: lamic orexin neurons in reward processing and addiction. Neuropharmacology 117–31. 2009;56(Suppl 1):112–21. [34] Marchant NJ, Densmore VS, Osborne PB. Coexpression of prodynorphin and [5] Zhou Y, Bendor J, Hofmann L, Randesi M, Ho A, Kreek MJ. Mu opioid receptor and corticotrophin-releasing hormone in the rat central amygdala: evidence of two orexin/hypocretin mRNA levels in the lateral hypothalamus and striatum are distinct endogenous opioid systems in the lateral division. J Comp Neurol 2007;504: enhanced by morphine withdrawal. J Endocrinol 2006;191:137–45. 702–15. [6] Sharf R, Sarhan M, Dileone RJ. Orexin mediates the expression of precipitated [35] Li Y, Li S, Wei C, Wang H, Sui N, Kirouac GJ. Changes in emotional behavior produced morphine withdrawal and concurrent activation of the nucleus accumbens shell. by orexin microinjections in the paraventricular nucleus of the thalamus. Pharmacol Biol Psychiatry 2008;64:175–83. Biochem Behav 2010;95:121–8. [7] Georgescu D, Zachariou V, Barrot M, Mieda M, Willie JT, Eisch AJ, et al. Involvement [36] Beck CH, Fibiger HC. Conditioned fear-induced changes in behavior and in the of the lateral hypothalamic peptide orexin in morphine dependence and expression of the immediate early gene c-fos: with and without diazepam withdrawal. J Neurosci 2003;23:3106–11. pretreatment. J Neurosci 1995;15:709–20. [8] Kirouac GJ, Parsons MP, Li S. Orexin (hypocretin) innervation of the paraven- [37] Igelstrom KM, Herbison AE, Hyland BI. Enhanced c-Fos expression in superior tricular nucleus of the thalamus. Brain Res 2005;1059:179–88. colliculus, paraventricular thalamus and septum during learning of cue-reward [9] Parsons MP, Li S, Kirouac GJ. The paraventricular nucleus of the thalamus as an association. Neuroscience 2010;168:706–14. interface between the orexin and CART peptides and the shell of the nucleus [38] Yasoshima Y, Scott TR, Yamamoto T. Differential activation of anterior and midline accumbens. Synapse 2006;59:480–90. thalamic nuclei following retrieval of aversively motivated learning tasks. [10] Li S, Kirouac GJ. Projections from the paraventricular nucleus of the thalamus to Neuroscience 2007;146:922–30. the forebrain, with special emphasis on the extended amygdala. J Comp Neurol [39] Brown EE, Robertson GS, Fibiger HC. Evidence for conditional neuronal activation 2008;506:263–87. following exposure to a cocaine-paired environment: role of forebrain limbic [11] Vertes RP, Hoover WB. Projections of the paraventricular and paratenial nuclei of structures. J Neurosci 1992;12:4112–21. the dorsal midline thalamus in the rat. J Comp Neurol 2008;508:212–37. [40] Rhodes JS, Ryabinin AE, Crabbe JC. Patterns of brain activation associated with [12] Hsu DT, Price JL. Paraventricular thalamic nucleus: subcortical connections and contextual conditioning to methamphetamine in mice. Behav Neurosci 2005;119: innervation by serotonin, orexin, and corticotropin-releasing hormone in 759–71. macaque monkeys. J Comp Neurol 2009;512:825–48. [41] Dayas CV, McGranahan TM, Martin-Fardon R, Weiss F. Stimuli linked to ethanol [13] Parsons MP, Li S, Kirouac GJ. Functional and anatomical connection between the availability activate hypothalamic CART and orexin neurons in a reinstatement paraventricular nucleus of the thalamus and dopamine fibers of the nucleus model of relapse. Biol Psychiatry 2008;63:152–7. accumbens. J Comp Neurol 2007;500:1050–63. [42] Hamlin AS, Clemens KJ, Choi EA, McNally GP. Paraventricular thalamus mediates [14] Alheid GF. Extended amygdala and basal forebrain. Ann NY Acad Sci 2003;985: context-induced reinstatement (renewal) of extinguished reward seeking. Eur J 185–205. Neurosci 2009;29:802–12. [15] de Olmos JS, Beltramino CA, Alheid GF. Amygdala and extended amygdala of the [43] Marcus JN, Aschkenasi CJ, Lee CE, Chemelli RM, Saper CB, Yanagisawa M, et al. rat: a cytoarchitectonical, fibroarchitectonical, and chemoarchitectonical survey. Differential expression of orexin receptors 1 and 2 in the rat brain. J Comp Neurol In: Paxinos G, editor. The rat nervous system. Third edition. New York: Academic 2001;435:6–25. Press; 2004. p. 509–603. [44] Cluderay JE, Harrison DC, Hervieu GJ. Protein distribution of the orexin-2 receptor [16] Alheid GF, Heimer L. New perspectives in basal forebrain organization of special in the rat central nervous system. Regul Pept 2002;104:131–44. relevance for neuropsychiatric disorders: the striatopallidal, amygdaloid, and [45] Li Y, Li S, Wei C, Wang H, Sui N, Kirouac GJ. Orexins in the paraventricular nucleus corticopetal components of substantia innominata. Neuroscience 1988;27:1–39. of the thalamus mediate anxiety-like responses in rats. Psychopharmacol (Berl) [17] Alheid GF, de Olmos JS, Beltramino CA. Amygdala and extended amygdala. In: 2010;212:251–65. Paxinos G, editor. The rat nervous system. 2nd ed. San Diego: Academic; 1995. [46] Borgland SL, Chang SJ, Bowers MS, Thompson JL, Vittoz N, Floresco SB, et al. Orexin A/ p. 495–578. hypocretin-1 selectively promotes motivation for positive reinforcers. J Neurosci [18] Frenois F, Cador M, Caille S, Stinus L, Le Moine C. Neural correlates of the 2009;29:11215–25. motivational and somatic components of naloxone-precipitated morphine [47] Sharf R, Sarhan M, Brayton CE, Guarnieri DJ, Taylor JR, DiLeone RJ. Orexin signaling withdrawal. Eur J Neurosci 2002;16:1377–89. via the orexin 1 receptor mediates operant responding for food reinforcement. [19] Gracy KN, Dankiewicz LA, Koob GF. Opiate withdrawal-induced fos immunoreac- Biol Psychiatry 2010;67:753–60. tivity in the rat extended amygdala parallels the development of conditioned [48] Boutrel B, Cannella N, de Lecea L. The role of hypocretin in driving arousal and place aversion. Neuropsychopharmacology 2001;24:152–60. goal-oriented behaviors. Brain Res 2010;1314:103–11. [20] Veinante P, Stoeckel ME, Lasbennes F, Freund-Mercier MJ. c-Fos and peptide [49] Azizi H, Mirnajafi-Zadeh J, Rohampour K, Semnanian S. Antagonism of orexin type immunoreactivities in the central extended amygdala of morphine-dependent 1 receptors in the locus coeruleus attenuates signs of naloxone-precipitated rats after naloxone-precipitated withdrawal. Eur J Neurosci 2003;18:1295–305. morphine withdrawal in rats. Neurosci Lett 2010;255:255–9. [21] Jin C, Araki H, Nagata M, Shimosaka R, Shibata K, Suemaru K, et al. Expression of c-Fos [50] Kelsey JE, Arnold SR. Lesions of the dorsomedial amygdala, but not the nucleus in the rat central amygdala accompanies the acquisition but not expression of accumbens, reduce the aversiveness of morphine withdrawal in rats. Behav Neurosci conditioned place aversion induced by withdrawal from acute morphine depen- 1994;108:1119–27. dence. Behav Brain Res 2005;161:107–12. [51] Watanabe T, Yamamoto R, Maeda A, Nakagawa T, Minami M, Satoh M. Effects of [22] Faure A, Richard JM, Berridge KC. Desire and dread from the nucleus accumbens: excitotoxic lesions of the central or basolateral nucleus of the amygdala on naloxone- cortical glutamate and subcortical GABA differentially generate motivation and precipitated withdrawal-induced conditioned place aversion in morphine-depen- hedonic impact in the rat. PLoS ONE 2010;5:e11223. dent rats. Brain Res 2002;958:423–8. [23] Lang PJ, Davis M. Emotion, motivation, and the brain: reflex foundations in animal [52] Nakagawa T, Yamamoto R, Fujio M, Suzuki Y, Minami M, Satoh M, et al. Involvement and human research. Prog Brain Res 2006;156:3–29. of the bed nucleus of the stria terminalis activated by the central nucleus of the 50 Y. Li et al. / Physiology & Behavior 102 (2011) 42–50

amygdala in the negative affective component of morphine withdrawal in rats. [59] Li Y, Li S, Sui N, Kirouac GJ. Orexin-A acts on the paraventricular nucleus of the Neuroscience 2005;134:9–19. midline thalamus to inhibit locomotor activity in rats. Pharmacol Biochem Behav [53] Valverde O, Tzavara E, Hanoune J, Roques BP, Maldonado R. Protein kinases in the rat 2009;93:506–14. nucleus accumbens are involved in the aversive component of opiate withdrawal. [60] Trivedi P, Yu H, MacNeil DJ, Van der Ploeg LH, Guan XM. Distribution of orexin Eur J Neurosci 1996;8:2671–8. receptor mRNA in the rat brain. FEBS Lett 1998;438:71–5. [54] Watanabe T, Nakagawa T, Yamamoto R, Maeda A, Minami M, Satoh M. [61] Hervieu GJ, Cluderay JE, Harrison DC, Roberts JC, Leslie RA. Gene expression and Involvement of noradrenergic system within the central nucleus of the amygdala protein distribution of the orexin-1 receptor in the rat brain and spinal cord. in naloxone-precipitated morphine withdrawal-induced conditioned place Neuroscience 2001;103:777–97. aversion in rats. Psychopharmacol (Berl) 2003;170:80–8. [62] Bayer L, Eggermann E, Saint-Mleux B, Machard D, Jones BE, Muhlethaler M, et al. [55] Delfs JM, Zhu Y, Druhan JP, Aston-Jones G. Noradrenaline in the ventral forebrain is Selective action of orexin (hypocretin) on nonspeciﬁc thalamocortical projection critical for opiate withdrawal-induced aversion. Nature 2000;403:430–4. neurons. J Neurosci 2002;22:7835–9. [56] Hou YY, Lu B, Li M, Liu Y, Chen J, Chi ZQ, et al. Involvement of actin rearrangements [63] Huang H, Ghosh P, van den Pol AN. Prefrontal cortex-projecting glutamatergic within the amygdala and the dorsal hippocampus in aversive memories of drug thalamic paraventricular nucleus-excited by hypocretin: a feedforward circuit withdrawal in acute morphine-dependent rats. J Neurosci 2009;29:12,244–54. that may enhance cognitive arousal. J Neurophysiol 2006;95:1656–68. [57] Carlezon Jr WA, Thomas MJ. Biological substrates of reward and aversion: a [64] Kolaj M, Doroshenko P, Yan Cao X, Coderre E, Renaud LP. Orexin-induced nucleus accumbens activity hypothesis. Neuropharmacology 2009;56(Suppl 1): modulation of state-dependent intrinsic properties in thalamic paraventricular 122–32. nucleus neurons attenuates action potential patterning and frequency. Neurosci- [58] Cardinal RN, Parkinson JA, Hall J, Everitt BJ. Emotion and motivation: the role ence 2007;147:1066–75. of the amygdala, ventral striatum, and prefrontal cortex. Neurosci Biobehav Rev [65] Sharf R, Sarhan M, Dileone RJ. Role of orexin/hypocretin in dependence and 2002;26:321–52. addiction. Brain Res 2010;1314:130–8.