Physiology & Behavior 109 (2013) 1–7

Contents lists available at SciVerse ScienceDirect

Physiology & Behavior

j ourna l homepage: www.e lsev ie r .com/ locate /phb

The effect of unpredictable chronic mild stress on depressive-like behavior and onhippocampal A1 and striatal A2A adenosine receptors

Leonardo M. Crema a,b, Letícia F. Pettenuzzo a,c, Michele Schlabitz a, Luisa Diehl a,b, Juliana Hoppe a,c,Régis Mestriner a,d, Daniela Laureano a,b, Christianne Salbego a,c, Carla Dalmaz a,b,c, Deusa Vendite a,c,⁎a Departamento de Bioquímica, ICBS, Universidade Federal do Rio Grande do Sul, Brazilb Programa de Pós-Graduação em Neurociências, ICBS, Universidade Federal do Rio Grande do Sul, Brazilc Programa de Pós-Graduação em Bioquímica, ICBS, Universidade Federal do Rio Grande do Sul, Brazild Programa de Pós-Graduação em Fisiologia, ICBS, Universidade Federal do Rio Grande do Sul, Brazil

H I G H L I G H T S

► UCMS rats demonstrated depressive-related behaviors.► UCMS and CRS rats presented increased A1R in the hippocampus.► UCMS rats presented increased A2AR in the striatum.► Upregulation of A2AR following UCMS may be associated with depressive behavior.

⁎ Corresponding author at: Departamento de Bioquímic2600, Anexo, 90035-003 Porto Alegre, RS, Brazil. Tel./fax:

E-mail address: vendite.ez@terra.com.br (D. Vendite

0031-9384/$ – see front matter © 2012 Published by Elhttp://dx.doi.org/10.1016/j.physbeh.2012.11.001

a b s t r a c t

a r t i c l e i n f o

Article history:Received 12 July 2012Received in revised form 16 September 2012Accepted 8 November 2012

Keywords:Chronic stressDepressionHippocampusStriatumAdenosineA1 receptorA2A receptorBindingWestern blotting

This study examined the effects of two chronic stress regimens upon depressive-like behavior, A1 and A2A

adenosine receptor binding and immunocontent. Male rats were subjected to unpredictable chronic mildstress (UCMS) or to chronic restraint stress (CRS) for 40 days. Subsequently, depressive-like behaviors (forcedswimming and consumption of sucrose) were evaluated, and A1 adenosine or A2A adenosine receptors were ex-amined in the hippocampus or striatum, respectively. UCMS animals demonstrated depressive-related behaviors(decrease in sucrose consumption and increased immobility in the forced swimming test). This group alsopresented increased A1 adenosine receptor binding and immunoreactivity in hippocampus, as well as increasedstriatal A2A adenosine receptor binding in the striatum, without alteration in immunoreactivity. Conversely,the chronic restraint stress group displayed only an increase in A1 adenosine receptor binding and no alterationin the other parameters evaluated. We suggest that the alteration in adenosine receptors, particularly theupregulation of striatal A2A adenosine receptors following UCMS, could be associated with depressive-relatedbehavior.

© 2012 Published by Elsevier Inc.

1. Introduction

Depression is a serious disorder, often manifested by symptoms atthe psychological, behavioral and physiological levels. Numerous at-tempts have been made develop animal models of depression or atleast of some aspects of the disease [1,2]. Most of these animal modelsshare the common feature of stress in the form of various stress pro-cedures or even aversive events, and chronic stress models appearmore suitable for the experimental investigation of depression thanacute stress models [3,4].

a, ICBS, UFRGS, Ramiro Barcelos,+55 51 3308 5570.).

sevier Inc.

Prolonged exposure of experimental animals to a variety of mildstressors is associated with significant changes in animal behavior. Theunpredictable chronicmild stress (UCMS)modelwas developed as an an-imal model of depression [4], and this animal model has also been associ-ated with the induction of “behavioral despair” in the forced swimmingtest (FST) [5]. In addition, one of the characteristics observed in animalmodels of depression is decreased consumption of and preference for pal-atable sweet solutions, indicating decreased responsiveness to rewardingstimuli [4,6]. UCMS simulates one main symptom of the melancholicsubtype of major depression, which involves several brain structures in-volved in themodulation of depressive behaviors, such as the hippocam-pus, amygdala, hypothalamus and prefrontal cortex [7–9]. Recent studieshave suggested functional and biochemical alterations in the striatum, in-duced by chronic stress, indicating that this structure is involved in themodulation of depressive behaviors [10,11]. On the other hand, we have

Table 1Schedule of stressor agents used during unpredictable chronic mild stress.

Days Stressors Days Stressors

1 1.5 h of restraint at 4 °C 21 2 h of flashing light2 Light during the night 22 2 h cage placed at 4 °C3 5 h of flashing light 23 24 h of isolation4 2 h of restraint 24 24 h of isolation5 4 h of cage placed at 4 °C 25 24 h of isolation6 24 h of isolation 26 1 h of restraint7 24 h of isolation 27 24 h of damp sawdust8 4 h of home cage inclination 28 3.5 h of flashing light9 1.5 h of restraint 29 3 h of home cage inclination10 2 h of flashing light 30 2 h of restraint at 4 °C11 Light during de night 31 No stressor applied12 3 h cage placed at 4 °C 32 24 h of food deprivation13 24 h of isolation 33 3 h of restraint14 24 h of isolation 34 4 h of flashing light15 No stressor applied 35 24 h of food deprivation16 12 h of damp sawdust 36 12 h of damp sawdust17 3 h of restraint 37 24 h of isolation18 6 h of home cage inclination 38 24 h of isolation19 2 h of restraint at 4 °C 39 Light during the night20 Light during the night 40 20 h of damp sawdust

2 L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

previously observed that animals submitted to chronic restraint stress, adifferent model of chronic stress, show increased ingestion of sweetfood. This alteration is probably due to a higher level of anxiety, becauseit was reversed by the administration of diazepam [12].

In the present study, we focused on the hippocampal A1 and striatalA2A receptors of the adenosinergic system. Adenosine is a brainneuromodulator and its extracellular concentration is increased in nox-ious brain conditions [13], namely during stress [14]. Its analogs havebeen shown to induce “behavioral despair” in animalmodels of depres-sion [15]. It can either inhibit or facilitate neuronal activity by activationof metabotropic A1 receptors or A2A receptors, respectively [13], both ofwhich are predominantly located in synapses in the limbic regions andneocortex [16,17]. However, the regulation of adenosine receptors isdynamic and is known to bemodified by chronic brain insults [18]. Fur-thermore, it is well established that the blockage of A2A receptorsaffords a robust neuroprotection in chronic noxious conditions in theadult brain [18]. Interestingly, A2A adenosine receptor antagonistshave been proposed as potential anti-depressants [15,19]. This is in ac-cordance with the ability of adenosine to control corticotrophin andcortisol/corticosterone release [20–22].

Given that glucocorticoids modify the expression of adenosine re-ceptors [23,24], we investigated the consequences of chronic restraintstress and unpredictable chronic mild stress on the modulation of A1

and A2A adenosine receptors in the hippocampus and striatum, re-spectively. We analyzed both UCMS and the chronic restraint stressparadigm to compare possible differences between these models ofchronic stress, since only UCMS is considered to be a model of depres-sion. Depressive-like behavior was also evaluated using the forcedswimming test (learned helplessness) and sucrose solution intake(decreased responsiveness to rewarding stimuli), respectively.

2. Experimental procedures

2.1. Animals

Seventy-six adult male Wistar rats (60 days at the beginning ofthe treatment, weighing 150–190 g) were used. Experimental ani-mals were housed in groups of 4–5 in home cages made of Plexiglasmaterial (65×25×15 cm) with the floor covered with sawdust andmaintained on a standard dark–light cycle (lights on between 7 and19 h), at a room temperature of 22±1 °C. The rats had free accessto food (standard lab rat chow) and water, except during the periodswhen restraint stress and water deprivation were applied. All animaltreatments were in accordance with the institutional guidelines andwere approved by the ethical committee; all efforts were made to re-duce the number of animals.

2.2. Stress models

The animals were divided into three groups: Control, Chronic Re-straint Stress (CRS) and Unpredictable Chronic Mild Stress (UCMS).Controls were kept undisturbed in their home cages during the entireperiod of treatment, receiving only ordinary facility care with dailysupport of food and water. Chronic restraint stress was applied byplacing the animal in a 25×7 cm plastic tube, and fixing it with plas-ter tape on the outside so that the animal was unable to move. Therewas a 1.5 cm hole at one far end for breathing. The restraint proce-dure was performed between 12 and 14 h. The animals were stressedfor 1 h/day, 5 days a week for 40 days [25]. The unpredictable chronicmild stress protocol was adapted from Gamaro et al. (1998) [25]. The40-day unpredictable chronic mild stress paradigm was used for theanimals in the UCMS group (Table 1). One of the following stressorswere used each day: (1) 24 h of damp sawdust; (2) light during thenight; (3) 1–3 h of restraint as described above; (4) 1.5–2 h of restraintat 4 °C; (5) 3–6 h of home cage inclination at an angle of 45°; (6) 2–5 hof flashing light, as described below; (7) isolation (2–3 days); (8) 1–4 h

of cage placed at 4 °C. Stress was applied at different schedules everyday, in order to increase unpredictability. Exposure to flashing lightwas made by placing the animal in an environment with a 40-W lampthat flashed at a frequency of 60 flashes/min.

2.3. Depressive-like behavior evaluation

Forced swimming test: 24 h after the last session of chronic stress,animals were placed individually in Plexiglas cylinders (height of40 cm, diameter of 18 cm) filled with water at 25±1 °C. Two differ-ent sessions were performed: a 15 min habituation session and a5 min test, 24 h later. The time of immobility was determined whenno additional activity was observed other than the movements neces-sary to keep the rat's head above the water [26]. Swimming was con-sidered when the rat showed active swimming movements, e.g.,moving around in the cylinder.

Sucrose solution intake: The sucrose consumption testwas performedonly in the UCMS group. Rats were trained to consume 1% (w/v) sucrosesolution before starting theUCMSprotocol. Habituation consisted of three1 h sessions (Monday,Wednesday and Friday), inwhich the animal couldselect between two bottles: 1% sucrose solution or tapwater. The sessionswere performed 20 h after food and water deprivation. During UCMS,sessions were conducted for 1 h, once a week (Wednesdays) during8 weeks, to evaluate preference [27]. During this period, the animalswere stressed daily, and the food and water removed the night beforethe test.

2.4. Preparation of total membranes from hippocampus and striatum

The rats were killed by decapitation, brains were removed and hippo-campus and striatum were dissected. Briefly, the two hippocampi andstriate from one animal were homogenized at 4 °C in a homogenizationbuffer containing a sucrose solution of 0.32 M, 50 mM Tris–HCl, 2 mMEGTA and 1 mM dithiothreitol, pH 7.6, centrifuged at 3000×g for10 min at 4 °C. The supernatants were collected, centrifuged at 14,000×g for 20 min at 4 °C and the pellet was resuspended at 4 °C in 500 μl in-cubation buffer containing 50 mM Tris–HCl (pH 7.4) and 2 mM MgCl2for the A1 adenosine receptor or 10 mMMgCl2 for the A2A adenosine re-ceptor (modified from Cunha et al.,1999) [28].

2.5. Binding studies

Hippocampal and striatal membranes were incubated with aden-osine deaminase (Calbiochem) for the A1 (2 U/ml) and A2A (4 U/ml)

3L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

adenosine receptors, respectively, in buffer containing 50 mM Tris–HCl, MgCl2 (2 mM for A1 and 10 mM for A2 receptors), pH 7.4, for30 min at 25 °C, in order to eliminate endogenous adenosine frommembrane preparations. Binding of the selective A1 receptor antago-nist, [3H]-1,3-dipropyl-8 cyclopentylxanthine (DPCPX) (specific ac-tivity of 120 Ci/mmol; from Perkin Elmer) was evaluated for 2 h at25 °C, with 40–100 μg of protein in a final volume of 300 μl in a solu-tion containing 50 mM Tris–HCl, 2 mM MgCl2, pH 7.4, as previouslydescribed by León et al. (2002) [29], with some modifications. A sat-uration curve was constructed using five different concentrations of[3H]-DPCPX (0.1, 0.5, 1, 5, and 10 nM). To evaluate the binding ofA2A receptors, we used a selective agonist, [3H]-2-[4-(2-p-carboxyethyl)phenylamino]-5′-N-ethylcarboxiamidoadenosine (CGS 21680) (40.5 Ci/mmol; from Perkin Elmer). Binding was measured at 25 °C for 4 h with30–100 μg of protein in a final volume of 300 μl solution containing50 mM Tris–HCl, 10 mM MgCl2, pH 7.4, as previously described byCunha et al. (1999) [28], with some modifications. A saturation curvewas constructed using five different concentrations of [3H]-CGS 21680(1, 2.5, 5, 10, and 20 nM). Specific bindingwas determined by subtractionof the non-specific binding, which was measured in the presence of aspecific concentration that was 10,000 times higher than that of 2-Cl-Adenosine (CADO) (from Research Biochemical Inc., Sigma-Aldrich).Each binding assay data point was performed in triplicate. The bindingreactions were stopped by rapid vacuum filtration through glass fiber fil-ters (GF/C filters), which were immediately washed three times with4 ml ice-cold buffer. The filterswere then placed in vials and 1 ml of scin-tillation liquid (from Perkin Elmer) was added. Radioactivity was deter-mined after at least 12 h with a counting efficiency of 55–60%. Thesampleswere counted for 2 min. The specific binding from saturation ex-periments was fitted by non-linear regression to a one binding site equa-tion using the software (GraphPad Inplot) to determine the bindingparameters (dissociation constant, Kd, and maximal number of bindingsites, Bmax).

2.6. Western blot analysis

For immunoblotting studies, hippocampal and striatal total mem-branes were prepared by homogenization in lysis buffer containing20 mMTris–HCl (pH 7.5), 150 mMNaCl, 1 mMEDTA, and 0.1% sodiumdodecyl-sulfate (SDS). Aliquots were taken for protein determinationand β-mercaptoethanol was added to a final concentration of 5%.Samples containing 50 mg of protein were resolved by 10% SDS-PAGE.After electrophoresis, proteins were electrotransferred to nitrocellulosemembranes using a semi-dry apparatus (Bio-Rad Trans-Blot SD). Themembranes were blocked for 1 h with 5% powdered milk in Tris-buffered saline plus 0.1% Tween-20, followed by incubation overnightat 4 °C with anti-A1 receptor (rabbit polyclonal antibody 119117,Calbiochem; 1:500) [30] or anti-A2A receptor (mouse monoclonal anti-body 05–717, Upstate; 1:1000) [31] diluted in the same blocking solu-tion. After washing, the membranes were incubated with adjustedsecondary antibodies coupled to horseradish peroxidase (Cell Signal-ing; 1:1000) for 2 h. All blots were re-probed with β-actin antibody(1:4000 dilution, mouse monoclonal, Sigma) as an internal control. Im-munoreactive bands were revealed by an enhanced chemilumines-cence kit (ECL Amersham from GE Healthcare), and detected usingX-rayfilms. The immunoblotfilmswere scanned and the digitalized im-ages analyzed with Optiquant software (Packard Instrument) [32].

Fig. 1. Effects of exposure to two chronic-stress models, chronic-restraint stress (CRS)or unpredictable chronic stress (UCMS), for 40 days on depressive-like behavior usingthe forced swimming task [one-way ANOVA showed a significant increase in theimmobility time (seconds) only for the UCMS group Pb0.05]. Data are expressed asmeans±standard error of the mean. N=23–24/group. *Significant difference fromcontrol and CRS groups (Pb0.05; Duncan multiple range test).

2.7. Protein assay

The total protein concentrations were determined using the meth-od described by Lowry et al. (1951) [33] for binding studies and byPeterson (1979) [34] for Western blot analysis, with bovine serum al-bumin as the standard.

2.8. Statistical analysis

Comparisons between experimental and control groups wereperformed by one way ANOVA followed by Duncan's multiple rangetest when appropriate and repeated measures ANOVA for sucroseconsumption test. A value of Pb0.05 was considered to be significant.

3. Results

In this study, only unpredictable chronic mild stress was able toincrease depressive-like behavior, evaluated as the time of immobili-ty in the forced swimming test, as shown in Fig. 1 [one-way ANOVA,F(2,68)=8.80, followed by Duncan's multiple range test, pb0.05].On the other hand, there was no significant difference between theCRS and control group for this task (p>0.05). Since our previous re-port has shown that animals subjected to CRS increase the consump-tion of sweet food [12], for the subsequent behavioral evaluation (thesucrose consumption task) we used only UCMS and control animals,which was applied to a different set of animals. There was no dif-ference among the groups in total fluid intake during the test sessions[repeated measures ANOVA, F(1,29)=1.251; P>0.1] (data notshown). As shown in Fig. 2, UCMS rats initiate the task by drinkingmore sucrose solution (first days of stress exposure) and diminishedthis consumption over time, in such a way that during the lastweeks of treatment they consumed significantly less sucrose solutionwhen compared to controls [repeated measures ANOVA [F(7,210)=6.03, pb0.05].

Rats submitted to UCMS or CRS presented an increase in Bmax forthe hippocampal A1 adenosine receptors, but no effect on Kd. Thesedata are shown in Fig. 3 [one-way ANOVA, Bmax: F(2,7)=6.42,pb0.05 and Kd: F(2,6)=0.51, p>0.05]. In addition, as evaluated byWestern Blotting analysis, both stress models (CRS and UCMS) wereable to increase A1 adenosine receptor immunoreactivity in this struc-ture, with UCMS showing a higher immunoreactivity than the CRSgroup, as displayed in Fig. 4 [one-way ANOVA, F(2,8)=18.5, followedby Duncan's multiple range test, pb0.05].

When we evaluated the striatal A2A adenosine receptors, only theUCMS group presented increased A2A receptor binding, but no effect onKd. These data are shown in Fig. 5 [one-way ANOVA, Bmax: F(2,14)=7.27, pb0.05 and Kd: F(2,15)=1.46, p>0.05]. However, a Westernblot analysis did not demonstrate any alteration in the immunoreactivityof striatal A2A adenosine receptors in any of the stressed groups[one-way ANOVA, F(2,6)=2.7, followed by Duncan's multiple rangetest, p>0.05], as shown in Fig. 6.

Fig. 2. Effects of exposure to unpredictable chronic stress (UCMS), for 40 days on thedepressive-like behavior, evaluating chronic sucrose consumption (repeated measuresANOVA showed a significant decrease in the sucrose intake in the UCMS group)(Pb0.05). Data are expressed as means±standard error of the mean. N=8/group.*Significant difference from control for the weeks numbers 7 and 8 (Pb0.05).

Fig. 4. Effects of exposure to two chronic stress models, chronic-restraint stress (CRS) orunpredictable chronic stress (UCMS), for 40 days on A1 adenosine receptor immunoreactiv-ity in the hippocampus. (A) A1 receptor immunoreactivity, (B) Data represent relative opticaldensity. A one-way ANOVA showed significant difference in the A1 receptor immunoreactiv-ity (Pb0.05). Data are expressed asmeans±standard error of themean for stressed groups/control group percentage. N=4–5/group. *Significant difference from control and UCMSgroup. **Significant difference from control and CRS groups (Pb0.05, Duncan multiplerange test).

4 L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

4. Discussion

In the present study, we observed an increase in the immobilitytime in the forced swimming task (FST) only in animals subjectedto the unpredictable mild chronic stress model (UCMS). Additionally,this group also showed reduced sucrose consumption at the end ofthe UCMS regimen, suggesting decreased responsiveness to reward-ing stimuli, a characteristic of depressive-like states in animal models.The UCMS group presented increased A1 adenosine receptor (A1Rs)binding (Bmax) and immunoreactivity in the hippocampus. Thisgroup also displayed increased A2A adenosine receptors (A2ARs) inthe striatum, as demonstrated by increases of Bmax, without alter-ation in the immunoreactivity. On the other hand, chronic restraintstress (CRS) only induced an increase in A1 adenosine receptor bind-ing and no alteration in the other measurements.

This is the first study to compare the effects of two different modelsof chronic stress, carried out at the same time on such parameters. Data

Fig. 3. Effects of exposure to two chronic-stress models, chronic-restraint stress (CRS)or unpredictable chronic stress (UCMS), for 40 days on A1 adenosine receptor bindingin the hippocampus. One-way ANOVA showed a significant increase in Bmax in bothstressed groups. There was no effect on the Kd in groups. Data are expressed as means±standard error of the mean. N=4–5/group. *Significant difference from control (Pb0.05,Duncan multiple range test).

from literature demonstrate that some models of unpredictable stress,such as themodels used herein, could be proposed asmodels of depres-sion in animal studies [4,35]. Although the FST is most widely recog-nized as a test for antidepressant drugs, the increased immobility inthe test session has received different interpretations. Behavioral

Fig. 5. Effects of exposure to two chronic stress models, chronic-restraint stress (CRS)and unpredictable chronic stress (UCMS), for 40 days on A2A receptor binding in thestriatum. A one-way ANOVA showed a significant increase in the A2A receptor Bmaxonly in the UCMS group. There was no effect on the Kd of groups. Data are expressedas means±standard error of the mean. N=4–5/group. *Significant difference fromcontrol and CRS groups (Pb0.05, Duncan multiple range test).

Fig. 6. Effects of exposure to two chronic stress models, chronic-restraint stress (CRS)or unpredictable chronic stress (UCMS), for 40 days on A2A receptor immunoreactivityin striatum. (A) A2A receptor immunoreactivity. (B) Data represent relative opticaldensity. There was no effect on A2A receptor immunoreactivity. Data are expressed asmeans±standard error of the mean for stressed groups/control group percentage.N=4–5/group (P>0.05, Duncan multiple range test).

5L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

immobility in the FST has usually been interpreted as “behavioral de-spair” [5]. Others have considered this immobility as a learned responsewhich reflects habituation to a stressful condition [36]. Although learn-ing may influence immobile behavior in the FST, UCMS animalspresented a higher immobility time than both the control and CRSgroups and presented diminished sucrose consumption. These dataare interpreted as “behavioral despair” and reduced response to re-warding stimuli, characterizing a depressive-like behavior [35,37–39].

Adenosine receptors A1 and A2a were assessed in hippocampus andstriatum, respectively. Structural and/or functional changes in the hippo-campus are associated with several mood disorders, includingmajor de-pression, and imaging studies have repeatedly emphasized the centralrole of the hippocampus in depression [40,41]. In addition, prolongedstress situations or exposure to high levels of glucocorticoids have beendemonstrated to induce hippocampal plasticity, causing decreased den-dritic arborization, spine production, and synaptic connectivity [42,43].The striatum plays a critical role in integrating sensory, emotional, mo-tivational andmotor components of ongoing actions [44]. Data from theliterature have suggested a principal role of this structure in depressive-like behavior [45,46], particularly with regard to responsiveness to re-warding stimuli.

One important finding of this studywas the observation of an altereddensity of hippocampal A1 adenosine receptor in both the chronic-stressmodels, as demonstrated by binding and Western blot studies, withoutany alteration in the receptor affinity (Kd). These data suggest that theincrease in these receptors is triggered by chronic stress rather than spe-cifically associated with the depressive-like state in rats. According toprevious investigations, the modification of the density of adenosinereceptors is related to the adaptation of the adenosine modulatory sys-tem under chronic noxious brain conditions [17,47]. A previous studyshowed that rats submitted to subchronic restraint stress for 7 daysshowed a down-regulation (i.e. decreased function) of A1 adenosine re-ceptor in the hippocampus [32]. This divergence with our results is

probably due to the type of stress used: 6 h of immobilization daily for7 days. Indeed, endogenous corticosteroids positively regulate the ex-pression of theA1 adenosine receptor in the rat brain [24], and,more spe-cifically, chronic restraint stress for 14 days increases the binding of theA1 adenosine receptor in the rat hypothalamus, demonstrating a possiblerelationship between the HPA axis and A1 adenosine receptor modula-tion [48]. Tagliari et al. (2010) [49] demonstrated that UCMS was ableto increase the levels of important proinflammatory cytokines, such asIL-1β, IL-6 and TNF-α, in the hippocampus of rats. In addition,interleukin-6 (IL-6) was reported to upregulate neuronal A1 adenosinereceptors. Such a mechanism could be involved in the enhancement ofA1 receptor-mediated signaling in the brain, with a beneficial impacton neuronal survival under excitotoxic situations [50]. Therefore, itmay be suggested that the up-regulation of adenosine A1 receptors,seen herein, could be involved in protecting hippocampal cells from in-sults induced by chronic stress.

Other important finding of the present study is the observation ofstriatal A2A adenosine receptor up-regulation in the UCMS group. How-ever, this up-regulation of A2A receptors, observed in the binding studies,was not followed by an increased A2A receptor immunoreactivity. Addi-tionally, CRS had no effect on striatal A2A adenosine receptors.

Both types of stress are known to release glucocorticoids [51,52], butonly UCMS showed changes in A2A adenosine receptor. These resultsappear consistent with published data that show that glucocorticoidsare unable to regulate A2A adenosine receptors [24]. A2A adenosine recep-tor modulation is complex, since A2ARs may form homomers and/orheteromers (including oligomers) with other receptors such as A1Rs, D2

dopamine receptors, metabotropic glutamate receptor 5 (mGlu5) andCB1 canabinoid receptors [53,54]. These receptor interactions could leadto modifications in the quaternary structure of A2ARs, acting as allostericeffectors [55]. In addition, A2ARs also interact with the so-called ‘accesso-ry’ proteins such asα-actinin, ARNO/cytohesin-2, USP4, NECAB2 and cal-modulin, and these interactions could also modulate the A2AR activity[56–58]. Thus, although the precise regulatory role of these interactionsremains to be established, the increased Bmax could be due to allostericactivation by interactions with other proteins.

Our results provide further support to the hypothesis that A2A re-ceptors in the striatum might be involved in depressive-like states; ithas been demonstrated that activation of A2A receptors in the ventralmedial striatum is a critical mediator of reserpine-induced depres-sion; furthermore a selective A2A antagonist, CSC, has been shownto ameliorate the performance of rats in the forced swimming test[59]. A recent paper has also reported that rats subjected to chronicunpredictable stress exhibited depressive-like behavior (decrease in su-crose consumption and increased immobility in the forced swimmingtest) and these effectswere reversed by chronic treatmentwith caffeine[60]. These and other studies [15,19,61] support a possible potential roleof A2A receptor antagonists as novel anti-depressants.

The mechanism by which A2A receptors antagonists act as antide-pressants probably involves interaction with D2 receptors at differentlevels. A2A receptors stimulate adenylyl-cyclase and activate thecAMP signaling pathway, with phosphorylation of several PKA sub-strates, such as DARPP-32, CREB and the consequent increase in theexpression of different genes. The activation of D2 receptor inhibitsthe effects of A2A receptor stimulation at the level of adenylyl cyclase.This antagonistic A2AR–D2R receptor interaction may also involve theformation of A2AR–D2R heterodimers, which modulates neuronal ex-citability and neurotransmitter release (for review see [62]).

In addition, the D2 receptor-like antagonist (haloperidol) was ableto prevent the antidepressant effect resulting from A2A receptorblockade or inactivation [15,61] and the actions of haloperidol oneffort-related actions were attenuated by antagonists A2aR [19]. How-ever, additional mechanisms such as the A2A receptor interactionwith other neurotransmitter systems or the ability of this receptorto control glial metabolism and neuroinflammation should also be ex-plored with future studies.

6 L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

4.1. Summary

In conclusion, both chronic stress models generated an up-regulationin hippocampal A1 adenosine receptors, while only unpredictable chronicmild stress (UCMS) promoted an up-regulation of A2A adenosine recep-tors in the striatum andwas capable of inducing depressive-like behavior.Although the effects, observed herein, on adenosine receptors and behav-ior suggest some correlation, it is not possible to draw causality fromthese results. Therefore, the relationship between striatal A2A adenosinereceptors and depressive-like behavior must be further studied.

Acknowledgments

This work was supported by the National Research Council of Brazil(CNPq), and PRONEX FAPERGS/CNPq 10/0018.3. Leonardo M. Cremawas the recipient of a CNPq fellowship.

References

[1] Cryan JF, Markou A, Lucki I. Assessing antidepressant activity in rodents: recentdevelopments and future needs. Trends Pharmacol Sci 2002;23:238-45.

[2] Holmes PV. Rodent models of depression: reexamining validity without anthro-pomorphic inference. Crit Rev Neurobiol 2003;15:143-74.

[3] Katz RJ, Roth KA, Carroll BJ. Acute and chronic stress effects on open field activityin the rat: implications for a model of depression. Neurosci Biobehav Rev 1981;5:247-51.

[4] Willner P. Animal models as simulations of depression. TIPS 1991;12:131-6.[5] Porsolt RD, Bertin A, Jalfre M. “Behavioural despair” in rats and mice: strain differences

and the effects of imipramine. Eur J Pharmacol 1978;51(3):291-4.[6] Katz RJ. Animal model of depression: pharmacological sensitivity of a hedonic

deficit. Pharmacol Biochem Behav 1982;16(6):965-8.[7] Kieran N, Ou XM, Iyo AH. Chronic social defeat downregulates the 5-HT1A receptor

but not Freud-1 or NUDR in the rat prefrontal cortex. Neurosci Lett 2010;469(3):380-4.

[8] Mozhui K, Karlsson RM, Kash TL, Ihne J, Norcross M, Patel S, et al. Strain differ-ences in stress responsivity are associated with divergent amygdala gene expres-sion and glutamate-mediated neuronal excitability. J Neurosci 2010;30(15):5357-67.

[9] Zambello E, Fuchs E, Abumaria N, Rygula R, Domenici E, Caberlotto L. Chronicpsychosocial stress alters NPY system: different effects in rat and tree shrew.Prog Neuropsychopharmacol Biol Psychiatry 2010;34:122-30.

[10] Dichter GS, Felder JN, Petty C, Bizzell J, Ernst M, Smoski MJ. The effects of psycho-therapy on neural responses to rewards in major depression. Biol Psychiatry2009;66(9):886-97.

[11] Marais L, Stein DJ, Daniels WM. Exercise increases BDNF levels in the striatum anddecreases depressive-like behavior in chronically stressed rats. Metab Brain Dis2009;24(4):587-97.

[12] Ely DR, Dapper V, Marasca J, Correa JB, Gamaro GD, Xavier MH, et al. Effect ofrestraint stress on feeding behavior of rats. Physiol Behav 1997;61:395-8.

[13] Fredholm BB, Chen JF, Cunha RA, Svenningsson P, Vaugeois JM. Adenosine andbrain function. Int Rev Neurobiol 2005;63:191-270.

[14] Scaccianoce S, Navarra D, Di Sciullo A, Angelucci L, Endröczi E. Adenosine andpituitary–adrenocortical axis activity in the rat. Neuroendocrinology 1989;50(4):464-8.

[15] El Yacoubi M, Costentin J, Vaugeois JM. Adenosine A2A receptors and depression.Neurology 2003;61(Suppl. 6):S82-7.

[16] Rebola N, Pinheiro PC, Oliveira CR, Malva JO, Cunha RA. Subcellular localization ofadenosine A(1) receptors in nerve terminals and synapses of the rat hippocam-pus. Brain Res 2003;987(1):49-58.

[17] Rebola N, Canas PM, Oliveira CR, Cunha RA. Different synaptic and subsynaptic lo-calization of adenosine A2A receptors in the hippocampus and striatum of the rat.Neuroscience 2005;132:893-903.

[18] Cunha RA. Neuroprotection by adenosine in the brain: From A(1) receptor activa-tion to A (2A) receptor blockade. Purinergic Signal 2005;1(2):111-34.

[19] Salamone JD, Farrar AM, Font L, Patel V, Schlar DE, Nunes EJ, et al. Differential ac-tions of adenosine A1 and A2A antagonists on the effort-related effects of dopa-mine D2 antagonism. Behav Brain Res 2009;201(1):216-22.

[20] Chau A, Rose JC, Koos BJ. Adenosine modulates corticotropin and cortisol releaseduring hypoxia in fetal sheep. Am J Obstet Gynecol 1999;180(5):1272-7.

[21] Geiger JD, Glavin GB. Adenosine receptor activation in brain reducesstress-induced ulcer formation. Eur J Pharmacol 1985;115(2–3):185-90.

[22] Jégou S, El Yacoubi M, Mounien L, Ledent C, Parmentier M, Costentin J, et al. Aden-osine A2A receptor gene disruption provokes marked changes in melanocortincontent and pro-opiomelanocortin gene expression. J Neuroendocrinol 2003;15(12):1171-7.

[23] Gerwins P, Fredholm BB. Glucocorticoid receptor activation leads to up-regulationof adenosine A1 receptors and down-regulation of adenosine A2 responses inDDT1 MF-2 smooth muscle cells. Mol Pharmacol 1991;40:149-55.

[24] Svenningsson P, Fredholm BB. Glucocorticoids regulate the expression of adeno-sine A1 but not A(2A) receptors in rat brain. J Pharmacol Exp Ther 1997;280(2):1094-101.

[25] Gamaro GD, Prediger ME, Lopes J, Bassani MG, Dalmaz C. Fluoxetine alters feedingbehavior and leptin levels in chronically-stressed rats. Pharmacol Biochem Behav2008;90(3):312-7.

[26] Porsolt RD, Bertin A, Blavet N, Deniel M, Jalfre M. Immobility induced by forcedswimming in rats: effects of agents which modify central catecholamine and sero-tonin activity. Eur J Pharmacol 1979;57:201-10.

[27] Moreau JL, Borgulya J, Jenck F, Martin JR. Tolcapone: a potential new antidepressantdetected in a novel animal model of depression. Behav Pharmacol 1994;5(3):344-50.

[28] Cunha RA, Constantino MD, Ribeiro JA. G Protein coupling of CGS 21680 bindingsites in the rat hippocampus and cortex is different from that of adenosine A 1and striatal A2A receptors. Naunyn Schmiedeberg's Arch Pharmacol 1999;359:295-302.

[29] León D, Albasanz JL, Ruíz MA, Fernández M, Martín M. Adenosine A1 receptordown-regulation in mothers and fetal brain after caffeine and theophylline treat-ments to pregnant rats. J Neurochem 2002;82:625-34.

[30] León D, Albasanz JL, Castillo CA, Martín M. Effect of glutamate intake during ges-tation on adenosine A(1) receptor/adenylyl cyclase pathway in both maternal andfetal rat brain. J Neurochem 2008;104:435-45.

[31] Marcellino D, Roberts DCS, Navarro G, Filip M, Agnati L, Lluis C, et al. Increase in A2Areceptors in the nucleus accumbens after extended cocaine self-administration andits disappearance after cocaine withdrawal. Brain Res 2007;1143:208-20.

[32] Cunha GM, Canas PM, Oliveira CR, Cunha RA. Increased density and synapto-protective effect of adenosine A2A receptors upon sub-chronic restraint stress.Neuroscience 2006;15:1775-81.

[33] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with theFolin phenol reagent. J Biol Chem 1951;193(1):265-75.

[34] Peterson GL. Review of the folin–phenol protein quantification method of Lowry,Rosebrough, Farr and Randall. Anal Biochem 1979;100:201-20.

[35] Pucilowski O, Overstreet DH, Rezvani AH, Janowsky DS. Chronic mild stress-induced anhedonia: greater effect in a genetic rat model of depression. PhysiolBehav 1993;54:1215-20.

[36] West AP. Neurobehavioral studies of forced swimming: the role of learning andmemory in the forced swim test. Prog Neuropsychopharmacol Biol Psychiatry1990;14:863-77.

[37] Quan MN, Zhang N, Wang YY, Zhang T, Yang Z. Possible antidepressant effects andmechanisms of memantine in behaviors and synaptic plasticity of a depression ratmodel. Neuroscience 2011;182:88-97.

[38] Taliaz D, Loya A, Gersner R, Haramati S, Chen A, Zangen A. Resilience to chronicstress is mediated by hippocampal brain-derived neurotrophic factor. J Neurosci2011;31(12):4475-83.

[39] Cox BM, Alsawah F, McNeill PC, Galloway MP, Perrine SA. Neurochemical, hor-monal, and behavioral effects of chronic unpredictable stress in the rat. BehavBrain Res 2011;220(1):106-11.

[40] Wang Y, Xiao Z, Liu X, Berk M. Venlafaxine modulates depression-induced behav-iour and the expression of Bax mRNA and Bcl-xl mRNA in both hippocampus andmyocardium. Hum Psychopharmacol 2011;26(2):95–101.

[41] Lucassen PJ, Stumpel MW, Wang Q, Aronica E. Decreased numbers of progenitorcells but no response to antidepressant drugs in the hippocampus of elderlydepressed patients. Neuropharmacology 2010;58(6):940-9.

[42] McKittrick CR, Magariños AM, Blanchard DC, Blanchard RJ, McEwen BS, Sakai RR.Chronic social stress reduces dendritic arbors in CA3 of hippocampus and de-creases binding to serotonin transporter sites. Synapse 2000;36(2):85-94.

[43] Fuchs E, Flugge G, Czeh B. Remodeling of neuronal networks by stress. Front Biosci2006;11:2746-58.

[44] Nestler EJ, Barrot M, DiLeone RJ, Eisch AJ, Gold SJ, Monteggia LM. Neurobiology ofdepression. Neuron 2002;34(1):13-25.

[45] Jahng JW. An animal model of eating disorders associated with stressful experi-ence in early life. Horm Behav 2011;59(2):213-20.

[46] Hollis F, Duclot F, Gunjan A, Kabbaj M. Individual differences in the effect of socialdefeat on anhedonia and histone acetylation in the rat hippocampus. Horm Behav2011;59(3):331-7.

[47] Coelho JE, Rebola N, Fragata I, Ribeiro JA, de Mendonça A, Cunha RA. Hypoxia-induced desensitization and internalization of adenosine A1 receptors in the rathippocampus. Neuroscience 2006;138(4):1195-203.

[48] Anderson SM, Leu JR, Kant GJ. Chronic stress increases the binding of the A1 aden-osine receptor agonist, [3H]cyclohexyladenosine, to rat hypothalamus. PharmacolBiochem Behav 1988;30(1):169-75.

[49] Tagliari B, Tagliari AP, Schmitz F, da Cunha AA, Dalmaz C, Wyse AT. Chronic vari-able stress alters inflammatory and cholinergic parameters in hippocampus ofrats. Neurochem Res 2011;36(3):487-93.

[50] Biber K, Pinto-Duarte A, Wittendorp MC, Dolga AM, Fernandes CC, Von FDKJ, et al.Interleukin-6 upregulates neuronal adenosine A1 receptors: implications forneuromodulation and neuroprotection. Neuropsychopharmacology 2008;33(11):2789-802.

[51] Bielajew C, Konkle ATM, Merali Z. The effects of chronic mild stress on maleSprague–Dawley and Long Evans rats: I. Biochemical and physiological analyses.Behav Brain Res 2002;136:583-92.

[52] Torres IL, Gamaro GD, Silveira-Cucco SN, Michalowski MB, Corrêa JB, Perry ML,et al. Effect of acute and repeated restraint stress on glucose oxidation to CO2 inhippocampal and cerebral cortex slices. Braz J Med Biol Res 2001;34:111-6.

[53] Kachroo A, Orlando LR, Grandy DK, Chen JF, Young AB, Schwarzschild MA. Inter-actions between metabotropic glutamate 5 and adenosine A2A receptors in nor-mal and Parkinsonian mice. J Neurosci 2005;25:10414-9.

[54] Martire A, Tebano MT, Chiodi V, Ferreira SG, Cunha RA, Köfalvi A, et al. Pre-synaptic adenosine A2A receptors control cannabinoid CB1 receptor-mediated

7L.M. Crema et al. / Physiology & Behavior 109 (2013) 1–7

inhibition of striatal glutamatergic neurotransmission. J Neurochem 2011;116(2):273-80.

[55] Gracia E, Pérez-Capote K, Moreno E, Barkešová J, Mallol J, Lluís C, et al. A2A aden-osine receptor ligand binding and signalling is allosterically modulated by adeno-sine deaminase. Biochem J 2011;435(3):701-9.

[56] Gsandtner I, Charalambous C, Stefan E, Ogris E, FreissmuthM, Zezula J. HeterotrimericG protein-independent signaling of a G-protein coupled receptor. Direct bindingof ARNO/cytohesin-2 to the carboxyl terminus of the A2A adenosine receptor isnecessary for sustained activation of the ERK/MAP kinase pathway. J Biol Chem2005;280(36):31898-905.

[57] Burgueño J, Blake DJ, BensonMA, Tinsley CL, Esapa CT, Canela EI, et al. The adenosineA2A receptor interacts with the actin-binding protein alpha-actinin. J Biol Chem2003;278:37545-52.

[58] Zezula J, Freissmuth M. The A(2A)-adenosine receptor: a GPCR with unique fea-tures? Br J Pharmacol 2008;153(Suppl. 1):S184-90.

[59] Huang QJ, Jiang H, Hao XL, Minor TR. Brain IL-1 beta was involved in reserpine-induced behavioral depression in rats. Acta Pharmacol Sin 2004;25(3):293-6.

[60] Pechlivanova MD, Tchekalarova JD, Alova LH, Petkov VV, Nikolov RP, YakimovaKS. Effect of long-term caffeine administration on depressive-like behavior inrats exposed to chronic unpredictable stress. Behav Pharmacol 2012;23:339-47.

[61] El Yacoubi M, Costentin J, Vaugeois JM. A2A receptors antagonists are potentialantidepressants: evidence based on pharmacology and A2A receptor knockoutmice. Br J Pharmacol 2001;134:68-77.

[62] Ferré S, Quiroz C, Woods AS, Cunha R, Popoli P, Ciruela F, et al. An update onadenosine A2A-dopamine D2 receptor interactions: implications for the functionof G protein-coupled receptors. Curr Pharm Des 2008;14(15):1468-74.