Changes in morphine-induced activation of cerebral Na+,K+-ATPase during morphine tolerance: Biochemical and behavioral consequences

Biochemical Pharmacology 83 (2012) 1572–1581

Changes in morphine-induced activation of cerebral Na+,K+-ATPase duringmorphine tolerance: Biochemical and behavioral consequences

Luis G. Gonzalez a,b,1, Willias Masocha c,1, Cristina Sanchez-Fernandez a,b, Ahmad Agil a,b,Maria Ocana a, Esperanza Del Pozo a,b,*, Jose M. Baeyens a,b

a Department of Pharmacology and Neurosciences Institute, Faculty of Medicine, University of Granada, Avenida de Madrid 11, 18012 Granada, Spainb Biomedical Research Center, University of Granada, Parque Tecnologico de Ciencias de la Salud, Armilla, 18100 Granada, Spainc Department of Applied Therapeutics, Faculty of Pharmacy, Kuwait University, P.O. Box 24923, Safat 13110, Kuwait

A R T I C L E I N F O

Article history:

Received 19 January 2012

Accepted 24 February 2012

Available online 3 March 2012

Keywords:

Analgesia

Morphine

Opioid receptors

Ouabain

Sodium–potassium ATPase

Tolerance

A B S T R A C T

There is ample evidence of the biological changes produced by the sustained activation of opioid

receptors. We evaluated the adaptive changes of cerebral Na+,K+-ATPase in response to the sustained

administration of morphine (minipumps, 45 mg/kg/day, 6 days) in CD-1 mice and the functional role of

these changes in opioid antinociception. The antinociceptive effect of morphine as determined with tail-

flick tests was reduced in morphine-tolerant mice. There were no significant changes in the density of

high-affinity Na+,K+-ATPase a subunits labeled with [3H]ouabain in forebrain membranes from

morphine-tolerant compared to those of morphine-naive animals. Western blot analysis showed that

there were no significant differences between groups in the changes in relative abundance of a1 and a3

subunits of Na+,K+-ATPase in the spinal cord or forebrain. However, the morphine-induced stimulation of

Na+,K+-ATPase activity was significantly lower in brain synaptosomes from morphine-tolerant mice

(EC50 = 1.79 � 0.10 mM) than in synaptosomes from morphine-naive mice (EC50 = 0.69 � 0.12 mM).

Furthermore, adaptive alterations in the time-course of basal Na+,K+-ATPase activity were observed after

sustained morphine treatment, with a change from a bi-exponential decay model (morphine-naive mice) to a

mono-exponential model (morphine-tolerant mice). In behavioral studies the antinociceptive effects of

morphine (s.c.) in the tail-flick test were dose-dependently antagonized by ouabain (1 and 10 ng/mouse,

i.c.v.) in morphine-naive mice, but not in morphine-tolerant mice. These findings suggest that during

morphine tolerance, adaptive cellular changes take place in cerebral Na+,K+-ATPase activity which are of

functional relevance for morphine-induced antinociception.

� 2012 Elsevier Inc. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Biochemical Pharmacology

jo u rn al h om epag e: ww w.els evier .c o m/lo cat e/bio c hem p har m

1. Introduction

Agonists of opioid receptors, including morphine, are essentialdrugs for pain alleviation. The repeated administration of opioidsleads to a progressive decrease in their potency [1], a phenomenonknown as tolerance, which represents a major problem related toopioid therapy. Adaptive cellular changes underlying opioidanalgesic tolerance include phosphorylation-regulated receptor

Abbreviations: ANOVA, analysis of variance; AUC, area under the curve; Bmax,

maximum number of binding sites; EC50, concentration of drug that produces half-

maximal stimulation; ED50, dose that produces 50% of maximum antinociception;

Emax, maximum efficacy; KD, dissociation constant; k�1, rate constant reduction of

enzymatic activity; t50, time required to reduce enzymatice activity by 50%; SEM,

standard error of the mean; TCA, trichloracetic acid.

* Corresponding author at: Department of Pharmacology and Neurosciences

Institute, Faculty of Medicine, University of Granada, Avda. Madrid 11, 18012

Granada, Spain. Tel.: +34 958 243539; fax: +34 958 243537.

E-mail address: [email protected] (E. Del Pozo).1 These authors contributed equally to this work.

0006-2952/$ – see front matter � 2012 Elsevier Inc. All rights reserved.

doi:10.1016/j.bcp.2012.02.023

internalization and trafficking [2], adaptations in adenylyl cyclasecAMP and protein kinase C signaling pathways [3], modifications inthe function and number of ion channels [4,5], changes in theexpression of several genes [6], and modifications in cellmorphology [7]. However, the mechanisms underlying opioidanalgesic tolerance have not yet been completely elucidated.

Pharmacological studies have shown that the acute activationof m opioid receptors modulates the activity of neuronal Na+,K+-ATPase in vitro through Gi/o protein-mediated mechanisms [8–10].Furthermore, the effect of morphine on Na+,K+-ATPase activityappears to be of functional relevance since ouabain and otherblockers of Na+,K+-ATPase are able to antagonize morphine-induced antinociception [11]. Na+,K+-ATPase comprises theenzymatic machinery involved in many aspects of neural activitysuch as restoring the ion gradient disturbed during electricalactivity, regulating the resting membrane potential and providingcation gradients that drive transmitter and metabolite uptakeprocesses [12]. Given its important role in regulating neuralexcitability, impairments in Na+,K+-ATPase activity might lead to

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–1581 1573

neural dysfunction [13,14]. In this connection, neurons frommorphine-tolerant guinea pigs were reported to show partialdepolarization of the cell membrane [15,16], which has beenhypothesized to result from an impairment in Na+,K+-ATPase[15,17]. In addition, downregulation of the abundance of synapticmembrane Na+,K+-ATPase in rats has been implicated in thesubsensitivity of neurons to opiates during the development oftolerance [18]. However, no published studies have attempted tocorrelate the changes during the development of morphinetolerance in cerebral Na+,K+-ATPase activity or density with thechanges in the analgesic effects of this drug.

Accordingly, we designed the present study to investigate theadaptive changes in synaptosomal Na+,K+-ATPase that take placeduring the sustained administration of morphine, by comparingthe stimulatory effect of morphine on Na+,K+-ATPase and the time-course of basal Na+,K+-ATPase activity in synaptosomes frommorphine-tolerant and morphine-naive mice. In addition, todetermine whether changes in the density of Na+,K+-ATPase pumpproteins after sustained morphine treatment were involved in thedevelopment of morphine tolerance, we tested the characteristicsof [3H]ouabain binding to neuronal membranes and used westernblotting to measure the abundance of a1 and a3 subunits ofNa+,K+,ATPase in forebrain and spinal cord membranes. Finally, weassessed the functional repercussions of the changes in Na+,K+-ATPase activity by evaluating the effect of ouabain (specificinhibitor of Na+,K+-ATPase) on morphine antinociception inmorphine-tolerant and morphine-naive mice.

2. Methods

2.1. Animals

Female CD-1 mice (Charles River, Barcelona, Spain), weighing25–30 g were used for all experiments. The animals were kept in atemperature-controlled room at 22 � 1 8C, with air exchange every20 min and an automatic 12 light/dark cycle (lights on from 08:00 to20:00 h). They were fed a standard laboratory diet and tap water ad

libitum until the beginning of the experiments. All experiments weredone during the same period of the day (09:00–15:00 h) to excludecircadian variations in the pharmacological effects.

The mice were handled in accordance with the EuropeanCommunities Council Directive of 24 November 1986 (86/609/ECC) for the care of laboratory animals. The experimental protocolwas approved by the Research Ethics Committee of the Universityof Granada, Spain.

2.2. Drugs and radioligands

Morphine hydrochloride was obtained from the GeneralDirectorate of Pharmacy and Drugs, Spanish Ministry of Health.Ouabain was obtained from Sigma–Aldrich Quimica SA (Madrid,Spain). [3H]ouabain (specific activity 16.5 Ci/mmol) was suppliedby NEM Life Science Products (Boston, MA, USA). The rest of thechemical products and reagents used in this study were obtainedfrom Sigma–Aldrich Quimica SA (Madrid, Spain).

2.3. Procedures

2.3.1. Experimental groups

The animals were treated with osmotic minipumps (Alzet 2001,Charles River, Barcelona, Spain) that released either morphine at arate of 45 mg/kg/day (morphine-tolerant group) or its vehicle(morphine-naive or control group) as described previously [5]. Theminipumps were implanted subcutaneously (s.c.) in animalsanesthetized with isoflurane. The mice were allowed to recoverfrom the anesthesia and given access to food and water until the

time of the experiment (6 days later). On the sixth day, the animals,with the pump still implanted, were used for antinociceptionexperiments or were killed and the forebrain and spinal cordmembranes were obtained as described below.

2.3.2. Preparation of forebrain P2 membranes

Mouse forebrain crude synaptosomal pellets were isolated aspreviously described [5]. Briefly, the mice were killed bydecapitation, the brains were quickly removed and the forebrainswere dissected and immersed in tubes containing ice-coldisolation medium I [320 mM sucrose; 3 mM ethylendiaminete-traacetic acid tetrasodium salt (EDTA�4 Na); 10 mM N-2-hydro-xyethylpiperazine-N0-ethanosulfonic acid (HEPES), pH 7.4]. Theneach forebrain was homogenized with three strokes of a Polytronhomogenizer (model PT10-35, Kinematica AG, Basel, Switzerland)set at position 3. Each stroke lasted 10 s and was separated fromthe next stroke by a 30 s stroke-free period during which the tubewas placed in ice. The homogenates were centrifuged (Avanti 30,Beckman Coulter Espana, SA, Madrid, Spain) at 1000 � g for10 min at 4 8C; the resulting pellets (containing nuclear and celldebris) were discarded and the supernatants were recentrifugedunder the same conditions. The final supernatant was thencentrifuged at 17 000 � g for 20 min to yield the crude synapto-somal pellet (P2 pellets). Then each pellet was resuspended eitherin 375 ml of medium I to isolate pure synaptosomes as describedbelow, or in the appropriate incubation medium for bindingexperiments.

2.3.3. Preparation of pure forebrain synaptosomes

Pure intact synaptosomes were obtained by Percoll densitygradient separation as previously described [19,20]. Percoll(Amersham Pharmacia Biotech, Madrid, Spain) stock solutionwas made by adding 0.5 ml of 2.5 M sucrose to 4.5 ml of originalPercoll. Solutions of lower Percoll concentration were prepared byappropriate dilution of the stock solution with medium II[250 mM sucrose; 10 mM HEPES; 3 mM EDTA�4 Na, pH 7.4]. Toprepare the Percoll density gradient, 3 ml of 16% Percoll solutionwas pipetted in the bottom of a 14 ml Ultra-Clear centrifuge tube,then 3 ml of 10% Percoll solution was layered over the 16% Percollsolution, and finally, 3.375 ml of a 7.5% Percoll solution (contain-ing 375 ml of the P2 pellet solution) was layered over the 10%Percoll solution. All steps were carried out at 4 8C. The tubes werecentrifuged at 15 000 � g for 20 min at 4 8C. Synaptosomesbanded at the 10%:16% Percoll interface were collected with awide-tip Pasteur pipette. To remove the Percoll from thesynaptosome preparations, the synaptosome/Percoll solutionwas dissolved (1:1, v/v) in a 320 mM sucrose solution andcentrifuged at 24 000 � g for 20 min at 4 8C. The supernatant wasdiscarded and the last centrifugation step was repeated. The finalpellet was dissolved in 1 ml of a 320 mM sucrose solution, and theprotein concentration was determined by a modified version ofthe Lowry method [21] using bovine serum albumin as thereference standard. After this, synaptosomes were diluted to therequired final protein concentration in medium III [320 nMsucrose; 10 mM HEPES, pH 7.4] and freshly made preparationswere used for ATPase assays or were stored at �20 8C. The storedsynaptosomes lost about 4% of their enzymatic activity after 15days (data not shown).

2.3.4. Na+,K+-ATPase assays

Na+,K+-ATPase activity was measured as previously described[8]. Briefly, to measure total Na+,K+-ATPase activity we added 50 mlof pure intact synaptosomes (final quantity in the assay medium0.01 mg protein) to a tube with 350 ml of an incubation mediumcontaining, in mM: 100 NaCl, 20 KCl, 2 MgCl2, 5 NaN3, 0.1 EGTA and25 HEPES, pH 7.4. The same medium but with 1 mM ouabain was

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–15811574

used to measure ouabain-insensitive ATPase. We then added 5 mlof morphine (or its vehicle) at concentrations from 1 nM to100 mM and preincubated the mixture for 5 min at 37 8C. After thistime the reaction was started by adding 50 ml of an ATP disodiumsalt solution (final ATP concentration in the medium 2 mM), andwas stopped 2 min later by adding 50 ml of 50% ice-coldtrichloracetic acid (TCA) and placing the tubes in ice for 10 min.The protocol of the time-course experiments to study basal Na+,K+-ATPase activity was slightly different, since morphine was notadded and different incubation times (from 0.5 to 60 min) with ATPwere allowed before adding TCA and placing the tubes in ice. Toremove protein precipitated by TCA, the sample was centrifuged at1000 � g for 10 min at 4 8C and 400 ml of the supernatant was usedto measure released inorganic phosphate (Pi) with the methodpreviously described [22,23]. Briefly, 400 ml of molybdate acidsolution color reagent was added to each tube (final volume800 ml), and after incubation for 30 min in the dark at roomtemperature, the absorbance was read at 810 nm with a microplatescanning spectrophotometer (PowerwaveX, Bio-Tek Instruments,Inc., Madrid, Spain). Sodium phosphate dibasic solution was usedas the reference standard.

At the end of the assay the amount of Pi in the presence ofouabain (ouabain-insensitive Na+,K+-ATPase activity) was sub-tracted from the amount of Pi in the absence of ouabain (totalNa+,K+-ATPase activity) to obtain the net Pi (NPi) produced byouabain-sensitive Na+,K+-ATPase. The values of net Pi (expressedas mM concentrations) were used to calculate the specific activity(SA) of the enzyme, according the following equation: SA [(nmolNPi/min) � mg�1] = [NPi (mM) � volume (ml) � (incubationtime)�1 � (mg membrane protein)�1]. To illustrate the enhance-ment of SA by morphine, the data were plotted as the percentageincrease in basal Na+,K+-ATPase activity (i.e. ATPase activitywithout morphine), which was calculated as follows: % increa-se = [(SA with morphine � SA without morphine)/SA withoutmorphine] � 100. The decay in basal SA with time was adjustedto an exponential decay model with nonlinear regression analysisfrom which the rate constants of the reduction in enzymaticactivity (k�1) and the time required to reach a 50% reduction inenzymatic activity (t50) were calculated (see Section 2.3.8).

2.3.5. Binding assays

The P2 pellet, obtained as described above (see Section 2.3.2),was dissolved in an incubation medium that contained, in mM, 150NaCl, 5 MgCl2, 1 EDTA�4 Na, 1.25 ATP and 50 Tris, pH 7.4.Membranes (50–60 mg/ml) were incubated in triplicate at 37 8C ina total volume of 500 ml with [3H]ouabain (1 nM, kinetics assays;0.5–64 nM, saturation assays) and 10 mM unlabeled ouabain (non-specific binding) or its solvent (total binding). The incubation timewas different depending on the type of assay. For associationassays, different times were used ranging from 0 to 120 min. Forprotein and saturation assays we used 90 min (the time necessaryto reach equilibrium; see Section 3.4). At the end of the incubationperiod, the reaction was stopped by adding 5 ml of an ice-coldsolution of the same composition as the incubation medium,except that ATP was not included. Each membrane solution wasimmediately filtered under a vacuum through Whatman GF/B glassfiber filters (SEMAT Technical Ltd., Banbury, UK) with a Brandel cellharvester (model M-12T, Brandel Instruments, Gaithersburg, MD,USA) and washed twice with 5 ml Tris HCl 50 mM, pH 7.4 at 4 8C. Indissociation assays we incubated membranes and radioligand for120 min; then unlabeled ouabain 10 mM (or its solvent) was addedand the reaction was stopped by rapid filtration at several times(120–210 min). The filters were transferred to scintillationcounting vials to which 4 ml of liquid scintillation cocktail(Optiphase Hisafe 2, Wallac Scintillation Products, London, UK)was added, and left to equilibrate in the dark for 12 h. The

radioactivity retained on the filters was measured with a liquidscintillation spectrometer (Beckman Instruments Espana, SA,Madrid, Spain) with an efficiency of 52%. Specific binding wascalculated by subtracting non-specific binding from total binding,and was linear up to a membrane protein concentration of 200 mg/ml (data from our lab not shown).

2.3.6. Western blot assays

Forebrain P2 fraction membranes were obtained as describedabove (see Section 2.3.2) with the exception of the homogenizationbuffer composition, which contained 10 mM Tris–HCl, pH 7.4,3 mM EDTA�4 Na and a protease inhibitor (Roche Diagnostics,Madrid, Spain; one tablet per 50 ml buffer, 6 ml added per mg oftissue). To obtain spinal cord membranes the vertebral column wassectioned at the thoracic and sacral levels and the spinal cord wasextracted by flushing 10 ml of ice-cold saline through the spinalcavity with a syringe. Then the spinal cords were immersed in thehomogenization buffer described above and subjected to the sameprocedure as the forebrain preparations to obtain the P2 fractionmembranes. The protein concentration was determined with themethod of Lowry [21]. After this, the P2 membranes were diluted tothe required final protein concentration and stored at �80 8C untiluse. Next, 20 ml of homogenate, corresponding to 20 mg of brain orspinal cord protein, was vigorously mixed with a 4 ml volume of 6�sample buffer (0.02% bromophenol blue, 6% mercaptoethanol, 40%glycerol, 8% SDS and 200 mM Tris–HCl, pH 6.8). This solution wasplaced in an Eppendorf tube and heated at 95 8C for 5 min.

Subsequently, samples were run on a 7.5% polyacrylamide gelat 120 V, 100 mA, during 100 min. Samples were next transferredonto a nitrocellulose membrane (Bio-Rad, Madrid, Spain) at roomtemperature (RT) during 30 min at 20 V and 100 mA. Thenitrocellulose membrane was blocked in 0.1% PBS-Tween with5% non-fat milk (blocking solution) for 60 min at RT. After three5 min washes in 0.1% PBS-Tween the membrane was incubatedovernight at 4 8C with the primary antibody against the a1 (1:100rabbit policlonal antibody, Abcam, Cambridge, UK) or a3 subunit(1:5000 mouse monoclonal antibody; Abcam) of Na+,K+-ATPase.The antibodies were diluted in blocking solution containing 0.5%non-fat milk. After incubation with Na+,K+-ATPase antibody, themembrane preparation was washed three times with 0.1% PBS-Tween and incubated with b-actin primary antibody (1:2 500mouse monoclonal antibody; Abcam, Cambridge, UK) for 60 min atRT. Then the membrane preparation was washed again with 0.1%PBS-Tween and incubated with the appropriate secondary anti-body (1:2 500 goat polyclonal secondary antibody to rabbit IgG forthe primary antibody against the a1 and 1:5000 goat polyclonalsecondary antibody to mouse IgG for the primary antibody againsta3 and b-actin; Abcam, Cambridge, UK) during 60 min at RT. Afterthe final washes, antibody binding was evaluated with anenhanced chemiluminescence method (ECL Plus western blottingdetection reagents from Amersham Biosciences, Buckinghamshire,UK) to detect immobilized specific antigens conjugated tohorseradish peroxidase-labeled antibodies, according to themanufacturer’s instructions. Immunoblots were analyzed bydensitometry using an instrument with reflectance capabilities(Kodak IS 4 000 MMPro, Carestream, Woodbridge, CT, USA) andMolecular Imaging Software (Carestream, Woodbridge, CT, USA).

2.3.7. Drug treatments and assessment of antinociception

Morphine was dissolved in ultrapure water and injected s.c. in avolume of 5 ml/kg. The Na+,K+-ATPase inhibitor ouabain wasdissolved in 1% Tween 80 in ultrapure water and injected i.c.v. in avolume of 5 ml/mouse. The control animals received the samevolume of vehicle. The s.c. injections were done in the inter-scapular region, and the i.c.v. injections were done in the rightlateral cerebral ventricle of mice according to the method

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–1581 1575

described previously [24]. Briefly, the injection site was identifiedaccording to the method of Haley and McCormick [25], and thedrug solution was injected with a 10 ml Hamilton syringe(Hamilton Company, Reno, Nevada, USA) with a sleeve aroundthe needle to prevent the latter from penetrating more than 3 mminto the skull. After the experiments were done, the position of theinjection was evaluated in each brain, and the results from animalsin which the tip of the needle did not reach the lateral ventriclewere discarded.

The antinociceptive effect of the treatments was evaluated witha tail-flick test as previously described [26]. Briefly, the animalswere restrained in a Plexiglas tube and placed on the tail-flickapparatus (LI 7100; Letica SA, Barcelona, Spain). A noxious beam oflight was focused on the tail about 4 cm from the tip, and thelatency to tail-flick was recorded automatically to the nearest 0.1 s.The intensity of the radiant heat source was adjusted to yieldbaseline latencies between 2 and 5 s; this intensity was neverchanged, and any animal whose baseline latency was outside thepre-established limits was excluded from the experiments. Twobaseline tail-flick latencies were recorded within 20 min before allinjections. Then the animals received an i.c.v. injection of ouabainor its solvent and immediately thereafter an s.c. injection ofmorphine or its solvent. The end of the last injection wasconsidered as time 0; from this time, tail-flick latencies weremeasured again at 10, 20, 30, 45, 60, 90, and 120 min aftertreatment. The cut-off time was 10 s.

The area under the curve (AUC) of tail-flick latency against timewas calculated for each animal with GraphPad Prism, 2007, v. 5.0software (GraphPad Software Inc., San Diego, CA, USA). The degreeof antinociception was determined according to the formula: %antinociception = [(AUCd � AUCv)/(AUCmax � AUCv)] � 100, wherethe AUCd and AUCv are the areas under the curve for drug- andvehicle-treated mice, respectively, and AUCmax is the area underthe curve of maximum possible antinociception (10 s in eachdetermination).

2.3.8. Data analysis

The parameters EC50 (concentration of morphine that producedhalf of the maximum enhancement of Na+,K+-ATPase activity),ED50 (dose of morphine that produced half of the maximalantinociception) and Emax (maximum increase in Na+,K+-ATPaseactivity or maximum antinociception produced) were calculatedfrom the concentration–response curves or dose–response curveswith nonlinear regression analysis (sigmoid curve, three param-eters) with the SigmaPlot, 2008, v. 11.0 program (SPSS Inc.,Chicago, IL, USA) and with GraphPad Prism, 2007, v. 5.0 software.

Fig. 1. Modulation of Na+,K+-ATPase activity by a single concentration of morphine as

synaptosomes were preincubated at 37 8C for different periods with 1 mM morphine o

5 min, and at time 0 of incubation the reaction was started with 2 mM ATP and then stopp

each point represents the mean � SEM of the values from four independent experiments

The enzymatic kinetic parameters of k�1 (rate constant of thereduction in enzymatic activity) and t50 (time required to reach a50% reduction in enzymatic activity) were also calculated with theSigmaPlot, 2008, v.11.0 program. The equilibrium dissociationconstant (KD) and the maximum number of binding sites (Bmax)from saturation binding assays were calculated and analyzed withthe KELL computer program for Windows, 1997, v. 6.0 (Biosoft,Cambridge, UK).

Mean values for two groups (western blot analysis) werecompared with Student’s t test. Mean values for more than twodifferent groups were compared with one- or two-way analysis ofvariance (ANOVA) followed by the Bonferroni post hoc test, usingGraphPad Prism, 2007, v. 5.0 and SigmaPlot, 2008, v. 11.0 software.The differences between means were considered significant whenthe value of P was below 0.05. The results in the text and figures areexpressed as the means � standard error of the mean (SEM).

3. Results

3.1. Optimal experimental conditions for evaluating the effect of

morphine on cerebral Na+,K+-ATPase activity in vitro

The experimental conditions used at our laboratory to evaluatethe effect of morphine on cerebral Na+,K+-ATPase activity, e.g. pH,temperature, and optimal protein concentration, have beendescribed previously [11]. In the present study, optimal pre-incubation or incubation times of the synaptosomes with a singleconcentration of morphine or ATP, respectively, were determinedanew in light of the diversity data in the literature regarding thesereaction parameters. As shown in Fig. 1A, the optimal preincuba-tion time of pure forebrain synaptosomes treated with either 1 mMmorphine or its vehicle was determined to be 5 min. After this timethe reaction was started with 2 mM ATP and then stopped atdifferent incubation times with 50% ice-cold trichloroacetic acid(Fig. 1B). Morphine (1 mM) stimulated Na+,K+-ATPase activity, butthis effect was inversely dependent on incubation time. Themaximum stimulatory effect of morphine on Na+,K+-ATPaseactivity was seen at 0.5 min, followed by a rapid decay. Stimulationremained evident during the first 10 min of incubation time, andthen disappeared completely at later times (Fig. 1B). At 2 minmorphine produced an increase of about 20% in Na+,K+-activitywith minimal dispersion of the data (Fig. 1B), and because this timeobviates the potential experimental difficulties inherent in shorterincubation times, it was used as the optimal incubation time forthe rest of the experiments.

a function of preincubation (A) and incubation time (B). In (A) control forebrain

r its solvent. In (B) synaptosomes were preincubated with 1 mM morphine during

ed after different incubation times with ice-cold trichloroacetic acid. In both graphs,

done in triplicate.

Fig. 2. Stimulatory effect of morphine on Na+,K+-ATPase activity in mouse forebrain

synaptosomes from morphine-naive and morphine-tolerant mice. Each point

represents the mean � SEM of the values from three to five independent experiments

(with different mice) done in triplicate. Statistically significant differences in

comparison to the effect of morphine on Na+K+-ATPase activity in synaptosomes

from morphine-naive mice: *P < 0.05, **P < 0.01 (two-way ANOVA followed by

Bonferroni post hoc test).

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–15811576

3.2. Morphine stimulates cerebral Na+,K+-ATPase activity in vitro and

this effect is decreased in morphine-tolerant mice

Different concentrations of morphine in vitro (1 nM to 100 mM)produced a concentration-dependent increase in ouabain-sensitiveNa+,K+-ATPase activity in forebrain synaptosomes from miceimplanted with vehicle-filled minipumps (Fig. 2). When thesynaptosomes were obtained from animals treated with minipumpsthat released morphine (morphine-tolerant mice, see Section 3.5) apartial but significant reduction in morphine-stimulated Na+,K+-ATPase activity was observed (Fig. 2). In this case, the shift in theconcentration–response curve was both rightward and downward.This indicated a significant increase in the EC50 of morphine from0.69 � 0.12 to 1.79 � 0.10 mM (P < 0.05). The Emax of morphinedecreased significantly from 24.05 � 0.12 to 19.55 � 0.10% (P < 0.01)in morphine-naive mice compared to morphine-tolerant mice (Fig. 2).

3.3. Morphine tolerance modulates the time-course of basal Na+,K+-

ATPase activity in mouse forebrain synaptosomes

The basal enzymatic activity of Na+,K+-ATPase in mouse forebrainmembranes decreased with time when the activity was evaluated

Fig. 3. Linearized plot for the time-course of basal Na+,K+-ATPase activity in mouse fore

from morphine-naive animals. (B) Monophasic decay of Na+,K+-ATPase activity in synapto

for the reduction in enzymatic activity with time. Each point represents the mean � SE

from 0 to 60 min in both morphine-naive and -tolerant mice;however, the decay exhibited different characteristics between thetwo types of synaptosomes (Fig. 3). In synaptosomes from morphine-naive animals, a detailed analysis of the time-course of the enzymaticactivity plot (nonlinear regression analysis) showed a better fit to abiphasic model than a monophasic model (P < 0.05, partial F test)with two rate constants for the reduction in activity:k�1A = 35.41 � 5.88 and k�1B = 2.19 � 0.14% � min�1 (Fig. 3A). Thetime required to reach 50% reduction (t50) in Na+,K+-ATPase enzymaticactivity was 2 min for the rapid reduction phase and 30 min for the slowreduction phase. In synaptosomes from animals rendered tolerant tomorphine, the time-course of Na+,K+-ATPase-specific activity showed amonophasic decay (r2 = 0.99). The single reduction rate constant (k�1)was 1.88 � 0.09% � min�1 and the t50 was 37 min (Fig. 3B).

3.4. Morphine tolerance does not modulate the density of cerebral

Na+,K+-ATPase a subunits in forebrain or whole spinal cord

Specific [3H]ouabain binding was linear at protein concentra-tions between 20 and 200 mg/ml (data not shown). Therefore,binding experiments were done at a final protein concentration of50–60 mg/ml. Kinetic binding assays for [3H]ouabain were done inmorphine-naive animals to determine the ligand incubation timefor equilibrium saturation assays. The steady state of association of[3H]ouabain was reached after 90 min at 37 8C (Fig. 4). Therefore,equilibrium saturation assays were done with an incubation timeof 90 min.

Saturation experiments showed that [3H]ouabain bound in asaturable manner to P2 brain membranes from morphine-naiveanimals (Fig. 5A) and morphine-tolerant mice (data not shown).The Scatchard analysis of these experiments yielded a monophasic(linear) plot over the concentration range tested in both types ofmembranes (Fig. 5B). The equilibrium dissociation constants (KD)were 14.73 � 1.58 for morphine-naive mice and 12.76 � 1.79 nM formorphine-tolerant mice, and the Bmax of [3H]ouabain-specific bindingsites was 28.12 � 3.69 pmol/mg protein in morphine-naive and33.14 � 5.61 pmol/mg protein in morphine-tolerant mice (Fig. 5B).Hill coefficients were not different from unity in membranes frommorphine-naive or morphine-tolerant mice (0.99 � 0.01 and1.00 � 0.02, respectively) (Fig. 5B, insert). These results suggest thepresence of only one class of binding site in the forebrain of bothgroups of animals, with similar (non-significantly differences) valuesof KD and Bmax. Considering that [3H]ouabain is a specific marker ofthe a subunits of Na+,K+-ATPase, no apparent change in the density ofa subunits of Na+,K+-ATPase in mouse forebrain membranes wasobserved after sustained treatment with morphine.

To verify the data generated by binding studies, two subunits(a1 and a3) of Na+,K+-ATPase were measured by western blot

brain synaptosomes. (A) Biphasic decay of Na+,K+-ATPase activity in synaptosomes

somes from morphine-tolerant animals. The k�1 values represent the rate constants

M of the values from four independent experiments done in triplicate.

Fig. 4. Kinetic binding assays of [3H]ouabain to morphine-naive mice forebrain membranes (P2 fraction). For association assays, membranes were incubated at 37 8C with

[3H]ouabain 1 nM for different periods ranging from 0 to 120 min. Once equilibrium was reached, dissociation was initiated by the addition of unlabeled ouabain (10 mM) and

was stopped by rapid filtration at the indicated times (120–210 min). The data for the time-course of both association and dissociation were fitted to a single-site binding

model (P < 0.01, partial F test). The inserts show the linear transformation of the association and dissociation data. The apparent association rate constant (kobs) and

dissociation rate constant (k�1) were 0.025 � 0.001 and 0.018 � 0.001 min�1, respectively. The k+1 (real association rate constant) was calculated as (kobs � k�1)/[Free], and the

calculated value of k+1 was 0.007 min�1 � nM�1. The data shown are representative of at least three experiments done in triplicate. The association was fitted to the equation:

Bt = (1 � e�kobst) and the data obtained from dissociation experiments were fitted to the equation: Bt = B0e�k�1t, where Bt is the amount of radioligand bound at time t; Be is the amount

of radioligand bound at equilibrium, and B0 is the amount of radioligand bound at time 0. The data shown are representative of at least three experiments done in triplicate.

Fig. 5. Assays of [3H]ouabain binding to mouse forebrain membranes (P2 fraction). (A) Representative saturation experiment in morphine-naive mice. Total, specific and non-

specific binding were plotted as a function of free ligand concentration. Non-specific binding represented less than 5% of the total binding. (B) Scatchard plots of specific

[3H]ouabain binding to forebrain membranes from both morphine-naive and morphine-tolerant mice. Monophasic plots were obtained in both experimental groups. The

insert shows the Hill plots of the data. Membrane proteins (50–60 mg/ml) and increasing concentrations of [3H]ouabain (0.5–64 nM) were incubated with 10 mM unlabeled

ouabain (non-specific binding) or its solvent (total binding) for 90 min at 37 8C. Specific binding was calculated as the difference between total and non-specific binding. Each

figure is representative of the results obtained in four experiments done in triplicate.

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–1581 1577

analysis. No significant changes in the relative abundance of a1 ora3 subunits of Na+,K+-ATPase were observed in forebrain or spinalcord membranes after sustained treatment with morphinecompared to morphine-naive animals (Fig. 6).

3.5. Morphine tolerance alters the antagonism by ouabain of the

antinociceptive effect of morphine

The administration of morphine (1–32 mg/kg, s.c.) togetherwith the ouabain vehicle (i.c.v.) produced a dose-dependentantinociceptive effect in morphine-naive mice (Fig. 7A). The ED50

of morphine calculated from the sigmoid curve yielded a value of1.98 � 0.14 mg/kg (Table 1). Treatment with ouabain (1 and 10 ng/mouse, i.c.v.) significantly reduced the antinociceptive effect ofmorphine. The dose–response curve of morphine was displaced to theright (Fig. 7A) and its ED50 was dose-dependently and significantly(P < 0.01) increased to 4.44 � 0.31 and 7.44 � 0.47 mg/kg formorphine plus ouabain at a dose of 1 and 10 ng/mouse, respectively

(Table 1). In morphine-tolerant mice, the s.c. administration ofmorphine (1–32 mg/kg) induced less antinociception than inmorphine-naive animals, with an ED50 of 4.19 � 0.21 mg/kg (mor-phine plus i.c.v. injection of the ouabain vehicle; Fig. 7B and Table 1).In contrast to morphine-naive animals, in morphine-tolerant animalsboth i.c.v. doses of ouabain failed (P > 0.05) to antagonize theantinociceptive effect of morphine, with an ED50 of 3.59 � 0.33 and3.78 � 0.25 mg/kg for morphine plus ouabain at 1 and 10 ng/mouse,respectively (Table 1 and Fig. 7B). There were no significantdifferences (P > 0.05) between Emax values obtained from mor-phine-naive and morphine-tolerant animals in the presence andabsence of ouabain (Table 1).

When the time-course of the antinociceptive effects wasplotted, we observed that in morphine-naive mice, both dosesof ouabain (1 and 10 ng/mouse, i.c.v.) significantly antagonized theantinociception produced by morphine (4 mg/kg, s.c.) from 10 to120 min after injection (Fig. 8A). When we compared thepercentages of antinociception calculated from changes with time

Fig. 6. Effects of sustained morphine treatment on the expression of a1 and a3

subunits of mouse forebrain and spinal cord Na+,K+-ATPase. (A) Representative

immunoblots for a1 and a3 subunits of Na+,K+-ATPase and b-actin; (B) relative

quantitative estimation (by scanning densitometry) of the expression of the

indicated protein. Equal quantities of protein (20 mg) prepared from solvent- and

morphine-treated mice were separated by polyacrylamide gel electrophophoresis

and then incubated with the primary antibody against the a1 or a3 subunits of

mouse Na+,K+-ATPase (1:100 and 1:500, respectively) overnight at 4 8C. The blots

were then washed and incubated for 60 min at room temperature with b-actin

primary antibody (1:2500) and subsequently with appropriate secondary

antibodies (1:2500) for 60 min at room temperature (see Section 2.3.6 for

details). Antibody binding was detected with an enhanced chemiluminescence

method according to the manufacturer’s instructions. Differences in the abundance

of each isoform of a subunits of Na+,K+-ATPase were found in homogenates from the

treatment groups (P > 0.05, Student’s t test). Sol, solvent; Mor, morphine. The data

shown are representative of five experiments done in duplicate.

Fig. 7. Effects of the association of ouabain (i.c.v.) or its solvent with morphine (s.c.)

morphine + ouabain 1 ng/mouse and morphine + ouabain 10 ng/mouse was evaluated in

antinociception was calculated from the area under the curve (AUC) of tail-flick latency

them represent the mean � SEM of the values obtained from 7 to 10 animals. Statistically s

way ANOVA followed by Bonferroni post hoc test).

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–15811578

in the AUC of antinociception, we found that ouabain (1 and 10 ng/mouse, i.c.v.) significantly (P < 0.01) antagonized the antinocicep-tive effect of morphine (4 mg/kg, s.c.) (Fig. 8B). In contrast, inanimals rendered tolerant to morphine neither of the doses ofouabain significantly modified the increase in tail-flick latencyinduced by morphine 4 mg/kg, s.c. at any time tested (Fig. 8C).Likewise, when the percentages of antinociception were analyzed,none of the doses of ouabain significantly modified the anti-nociceptive effect of morphine (4 mg/kg, s.c.) (Fig. 8D).

The administration of both doses of ouabain together with themorphine solvent (s.c.) did not significantly modify basal tail-flicklatency at any time or dose tested (data not shown).

4. Discussion

The results of this study show that sustained exposure tomorphine in vivo decreases its stimulatory effect on cerebralNa+,K+-ATPase activity and modifies the time-course of Na+,K+-ATPase activity in vitro. These changes have functional repercus-sions in vivo that affect the antagonism by ouabain of theantinociceptive effect of morphine.

Earlier reports showed that morphine increased Na+,K+-ATPaseactivity in the brain by activating m-opioid receptors, an effectsensitive to opioid receptor antagonism [8,10]. One suggestedmolecular model for the control of Na+,K+-ATPase by m-opioidreceptors postulated that acute m-opioid receptor activation by Gi/

o proteins inhibits adenylyl cyclase and decreases cAMP-depen-dent protein kinase (PKA) activation, leading to a decrease in thephosphorylation of Na+,K+-ATPase, which in turn enhances Na+,K+-ATPase activity [8,10,27]. Because the stimulation of Na+,K+-ATPase seems to play a role in the acute effect of morphine,modifications in the function of this enzyme would be expected aspart of the adaptive changes that take place during morphinetolerance. In fact, our results show that cerebral Na+,K+-ATPaseactivity in animals rendered tolerant to morphine decreased whencompared to that of morphine-naive animals. In agreement withour results, other authors have reported that long-term exposureto morphine in vivo is linked to impaired electrogenic activity ofNa+,K+-ATPase in the locus ceruleus and myenteric plexus of theguinea pig ileum [16,17]. In addition, an alteration in mousehippocampal Na+,K+-ATPase activity was shown after long-termopioid treatment, which seemed to be associated with upregula-tion of the cAMP/PKA signaling pathway [10].

The present study demonstrates for first time that the time-course of basal Na+,K+-ATPase activity changes from a biphasicdecay model in synaptosomes from morphine-naive animals to a

on the tail-flick test in mice. The antinociception induced by morphine + vehicle,

morphine-naive animals (A) and morphine-tolerant animals (B). The percentage of

with time (as described in Section 2.3.7). Each point and the vertical lines that cross

ignificant differences in comparison to morphine + vehicle: *P < 0.05, **P < 0.01 (two-

Table 1Parameters of dose–response curves for the antinociceptive effects of morphine + vehicle and morphine + ouabain 1 or 10 ng in morphine-naive and -tolerant mice.

Experimental groups Treatments

Morphine + vehicle Morphine + ouabain 1 ng Morphine + ouabain 10 ng

Morphine-naive ED50 (mg/kg) 1.98 � 0.14 4.44 � 0.31** 7.44 � 0.47**,§

Emax (% analgesia) 79.92 � 5.09 68.18 � 6.84 64.48 � 5.28

Morphine-tolerant ED50 (mg/kg) 4.19 � 0.21 3.59 � 0.33 3.78 � 0.25

Emax (% analgesia) 77.35 � 5.08 74.34 � 5.07 73.38 � 5.08

Antinociception was evaluated in the tail-flick test in mice and analyzed as percentage of maximal analgesia from AUC values of tail-flick latencies along time (see Section

2.3.7 for details). Morphine was administered s.c. and ouabain, i.c.v. ED50 and Emax values were calculated from dose–response curves using a nonlinear regression analysis,

and expressed as mean � SEM (n = 7–10 mice).** P < 0.01 compared with morphine + vehicle.§ P < 0.01 compared with morphine + ouabain 1 ng (one-way ANOVA followed by Bonferroni post hoc test).

Fig. 8. Antagonism by i.c.v. treatment with ouabain of the antinociception induced by morphine (4 mg/kg, s.c.) in a tail-flick test in morphine-naive (A and B) and -tolerant (C

and D) mice. Time-course of the tail-flick latency times for morphine + vehicle and morphine + ouabain (1 and 10 ng/mouse, i.c.v.) in morphine-naive- (A) and morphine-

tolerant (C) animals. Each point represents the mean � SEM of the values obtained from 7 to 10 animals. Statistically significant differences in comparison to morphine + vehicle:

*P < 0.05, **P < 0.01 (two-way ANOVA followed by Bonferroni post hoc test). Effects of ouabain (1–10 ng/mouse, i.c.v.) on morphine-induced antinociception in morphine

morphine-naive (B) and morphine-tolerant (D) animals. The solid column represents the effect of morphine + vehicle. The percentage of antinociception was calculated from the

area under the curve (AUC) of tail-flick latency with time (as described in Section 2.3.7). Each column represents the mean � SEM of the values obtained from 7 to 10 animals.

Statistically significant differences in comparison to morphine + vehicle: **P < 0.01 (one-way ANOVA followed by Bonferroni post hoc test).

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–1581 1579

monophasic model in synaptosomes from morphine-tolerantanimals. As previously mentioned, Na+,K+-ATPase activity maybe regulated by phosphorylation processes catalyzed by proteinkinases, and is negatively controlled by these phosphorylationprocesses [28,29]. This would explain the different time-courseprofiles for Na+,K+-ATPase activity observed in our two types ofbrain synaptosomes. In preparations from morphine-naive ani-mals, the addition of ATP to the reaction medium could triggerphosphorylation mechanisms of Na+,K+-ATPase, which could inturn lead to biphasic patterns of reduction in its activity throughindependent but parallel pathways for ATP hydrolysis. This idea issupported by other authors [30], who described a complex kineticbehavior of Na+,K+-ATPase activity secondary to multiphasicpatterns of phosphorylation and dephosphorylation catalyzed bymammalian Na+,K+-ATPases when ATP, Mg2+ and Na+ were addedto the reaction medium. We hypothesized that these and otherphosphorylation/dephosphorylation mechanisms of Na+,K+-ATPase may be altered after sustained morphine treatment, an

effect which may underlie the modifications in kinetic behaviorobserved in brain synaptosomes from morphine-tolerant micecompared to morphine-naive mice. Preliminary studies showedthat in control synaptosomes the stimulatory effect of morphine onNa+,K+-ATPase was observed only during the first 10 min ofincubation, coinciding with the first phase of the biphasic decay inNa+,K+-ATPase activity, but disappeared completely at later times(for more details see Fig. 1). These findings suggest that only thefirst phase of basal Na+,K+-ATPase activity is sensitive to thestimulatory effect of morphine, a phase that is lost under toleranceconditions. From a mechanistic standpoint this finding suggeststhat the presence of compensatory mechanisms resulting fromsustained exposure to morphine may be responsible for thedisappearance of the phase sensitive to acute morphine.

To determine whether changes in the density of Na+,K+-ATPaseafter sustained morphine treatment are involved in alterations inNa+,K+-ATPase activity, we tested the binding of [3H]ouabain, ahighly specific and well-characterized inhibitor of Na+,K+-ATPase

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–15811580

[31], to forebrain membranes from morphine-naive and -tolerantmice. The constituent a and b subunits of Na+,K+-ATPase areheterogenous [32]. With regard to the catalytic a subunit, threeisoforms have been described in the central nervous system (a1,a2, a3) with low (millimolar), intermediate (micromolar) and high(nanomolar) affinity for ouabain, respectively [32]. The analysis ofequilibrium binding assays, as well as kinetic binding assays,showed evidence for only one binding site with high affinity (KD inthe low nanomolar range) for the radioligand, which agrees withpreviously reported data [33]. Our results strongly suggest thatunder our experimental conditions, only the a3 subunit isoform ofNa+,K+-ATPase was identified. This proposal is based on theconsiderations that (1) only the affinity of the a3 subunit is strongenough for ouabain to be labeled by nanomolar concentrations ofthe radioligand, and (2) according to current knowledge about thetissue distribution of different a subunits, the a3 is found mainly inneurons [28]. When the animals were made tolerant, thecharacteristics of this high-affinity binding site of [3H]ouabainwere not altered compared to preparations from solvent-treatedanimals, which indicates that chronic exposure to morphine doesnot result in any substantial modification in the density of the a3

subunit of cerebral Na+,K+-ATPase. As a check for the informationobtained with [3H]ouabain binding assays, western blot analyseswere done with both forebrain and spinal cord samples. Theseassays confirmed the data generated by binding studies, andfurther showed that there were no changes in the abundance of a1

and a3 subunits of Na+,K+-ATPase in the forebrain or spinal cordunder tolerant conditions.

These results are supported by earlier findings that neithershort-term nor long-term morphine treatment is associated withany change in the abundance of specific a subunits (a1 and a3) ofNa+,K+-ATPase in the mouse hippocampus and striatum [10,34].However, our findings contrast with those of other authors whodemonstrated a reduction in the abundance of Na+,K+-ATPase in asynaptic-plasma membrane fraction from tolerant rats [18], and aspecific decrease in the density of the a3 subunit of Na+,K+-ATPasein guinea pig myenteric neurons after sustained exposure tomorphine [17,35]. Apparently, depending on the tissue studied andthe experimental conditions, the abundance of Na+,K+-ATPase maybe reduced or may remain unchanged.

In contrast to the absence of changes in the binding and westernblot experiments in morphine-tolerant animals, we observedsignificant differences in Na+,K+-ATPase activity between mor-phine-naive and -tolerant animals. This finding was not entirelyunexpected, because previous studies have shown Na+,K+-ATPaseactivity to be modulated by a phosphorylation process mediatedby PKA, with no change in the abundance of the enzyme measuredby [3H]ouabain binding [36] or western blot [34]. These observa-tions can be explained, in part, by two facts. Firstly, duringmorphine tolerance the activity of some protein kinases and thedegree of phosphorylation of several proteins are enhanced[37,38]. Secondly, the phosphorylation of specific subunits ofNa+,K+-ATPase is associated with inhibition of its activity [36,39].Therefore, phosphorylation of Na+,K+-ATPase induced by thesustained administration of morphine may lead to changes insome intrinsic properties of the enzyme, such as the E1–E2conformational equilibrium [40]. This would produce changes inenzyme activity such as those we found, but not in its abundance atthe cell surface. Nevertheless, we cannot rule out other possibili-ties such as the regulation, induced by morphine tolerance, of thespecific b subunits of Na+,K+-ATPase.

The antagonism of the antinociceptive effect of morphine bydigitalis glycosides was analyzed in detail in a previous publicationfrom this laboratory [11]. This effect cannot be explained by adirect interaction between morphine and digitalis glycosides attheir binding sites, because our previous findings showed that

ouabain did not modify the specific binding of [3H]naloxone fromthe opioid receptor, and morphine did not displace [3H]ouabain-specific binding [8,11].

We now show for the first time that ouabain dose-dependentlyantagonizes the antinociception induced by morphine in controlanimals but does not modify the antinociceptive effect ofmorphine in animals rendered tolerant. This finding is supportedby the work of Kong et al. [15], who reported that the depolarizingeffect of ouabain in guinea pig myenteric neurons was negligible inanimals chronically treated with morphine compared to theircontrols. The loss of effect of ouabain (which targets this ATPase)under tolerance condition can probably be explained by thealterations in Na+,K+-ATPase activity reported previously duringmorphine tolerance. Our in vivo results provide a functionalcorrelate to our findings for enzymatic activity and kinetics, andsuggest that during morphine tolerance, alterations in the ouabaintarget (Na+,K+-ATPase) were associated with both a decrease in thedegree of antinociception and the disappearance of the inhibitoryeffect of ouabain on morphine-induced antinociception.

In summary, this study shows that the sustained administrationof morphine to mice does not modify the abundance of Na+,K+-ATPase a subunits in the forebrain or spinal cord, but does inducechanges in the basal activity of cerebral Na+,K+-ATPase and reducesthe ability of morphine to stimulate it. The changes observed in

vitro have functional consequences in tolerant animals, in whichthe effect of ouabain on morphine antinociception is lost.

Acknowledgments

This study was supported by grants from the Junta de Andalucıa(CTS-109). C.S.F. was supported by an FPU-MED grant. The authorsthank K. Shashok for revising the English style of the manuscript.

References

[1] Ueda H, Ueda M. Mechanisms underlying morphine analgesic tolerance anddependence. Front Biosci 2009;14:5260–72.

[2] Zhang Y, Xiong W, Lin X, Ma X, Yu LC. Receptor trafficking induced by mu-opioid-receptor phosphorylation. Neurosci Biobehav Rev 2009;33:1192–7.

[3] Gintzler AR, Chakrabarti S. The ambiguities of opioid tolerance mechanisms:barriers to pain therapeutics or new pain therapeuctic possibilities. J Pharma-col Exp Ther 2008;325:709–13.

[4] Kennedy C, Henderson G. Mu-opioid receptor inhibition of calcium current:development of homologous tolerance in single SH-SY5Y cells after chronicexposure to morphine in vitro. Mol Pharmacol 1991;40:1000–5.

[5] Gonzalez LG, Portillo E, del Pozo E, Baeyens JM. Changes in [3H]glibenclamidebinding to mouse forebrain membranes during morphine tolerance. Eur JPharmacol 2001;418:29–37.

[6] Tapocik JD, Letwin N, Mayo CL, Frank B, Luu T, Achinike O, et al. Identification ofcandidate genes and gene networks specifically associated with analgesictolerance to morphine. J Neurosci 2009;29:5295–307.

[7] Mayer DJ, Mao J, Holt J, Price DD. Cellular mechanisms of neuropathic pain,morphine tolerance, and their interactions. Proc Natl Acad Sci USA 1999;96:7731–6.

[8] Masocha W, Gonzalez LG, Baeyens JM, Agil A. Mechanisms involved in mor-phine-induced activation of synaptosomal Na+,K+-ATPase. Brain Res 2002;957:311–9.

[9] Horvath G, Agil A, Joo G, Dobos I, Benedek G, Baeyens JM. Evaluation ofendomorphin-1 on the activity of Na+,K+-ATPase using in vitro and in vivostudies. Eur J Pharmacol 2003;458:291–7.

[10] Wu ZQ, Li M, Chen J, Chi ZQ, Liu JG. Involvement of cAMP/cAMP-dependentprotein kinase signaling pathway in regulation of Na+,K+-ATPase upon activa-tion of opioid receptors by morphine. Mol Pharmacol 2006;69:866–76.

[11] Masocha W, Horvath G, Agil A, Ocana M, del Pozo E, Szikszay M, et al. Role ofNa+,K+-ATPase in morphine-induced antinociception. J Pharmacol Exp Ther2003;306:1122–8.

[12] Aperia A. New roles for an old enzyme: Na+,K+-ATPase emerges as an inter-esting drug target. J Intern Med 2007;261:44–52.

[13] Clapcote SJ, Duffy S, Xie G, Kirshenbaum G, Bechard AR, Rodacker Schack V,et al. Mutation I810N in the a3 isoform of Na+,K+-ATPase causes impairmentsin the sodium pump and hyperexcitability in the CNS. Proc Natl Acad Sci USA2009;106:14085–90.

[14] Benarroch EE. Na+,K+-ATPase. Functions in the nervous system and involve-ment in neurologic disease. Neurology 2011;76:287–93.

L.G. Gonzalez et al. / Biochemical Pharmacology 83 (2012) 1572–1581 1581

[15] Kong JQ, Leedham JA, Taylor DA, Fleming WW. Evidence that tolerance anddependence of guinea pig myenteric neurons to opioids is a function of alteredelectrogenic sodium–potassium pumping. J Pharmacol Exp Ther 1997;280:593–9.

[16] Kong JQ, Meng J, Biser PS, Fleming WW, Taylor DA. Cellular depolarization ofneurons in the locus ceruleus region of the guinea pig associated withthe development of tolerance to opioids. J Pharmacol Exp Ther 2001;298:909–16.

[17] Biser PS, Thayne KA, Fleming WW, Taylor DA. Na+,K+-ATPase a-subunitisoform distribution and abundance in guinea-pig longitudinal muscle/myen-teric plexus after exposure to morphine. Brain Res 2002;931:186–93.

[18] Prokai L, Zharikova AD, Stevens Jr SM. Effect of chronic morphine exposure onthe synaptic plasma-membrane subproteome of rats: a quantitative proteinprofiling study based on isotope-coded affinity tags and liquid chromatogra-phy/mass spectrometry. J Mass Spectrom 2005;40:169–75.

[19] Nagy A, Delgado-Escueta AV. Rapid preparation of synaptosomes from mam-malian brain using nontoxic isoosmotic gradient material (Percoll). J Neuro-chem 1984;43:1114–23.

[20] Araya KA, Pessoa CD, Gonzalez LG. Role of CB1 receptors and Gi/o proteinactivation in the modulation of synaptosomal Na+,K+-ATPase activity byWIN55,212-2 and D9-THC. Eur J Pharmacol 2007;572:32–9.

[21] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with theFolin phenol reagent. J Biol Chem 1951;193:265–75.

[22] Fiske CH, Subbarow Y. The colorimetric determination of phosphorus. J BiolChem 1925;66:375–400.

[23] Sadrzadeh SM, Vincenzi FF, Hinds TR. Simultaneous measurement of multiplemembrane ATPases in microtiter plates. J Pharmacol Toxicol Methods 1993;30:103–10.

[24] Moncada A, Cendan CM, Baeyens JM, Del Pozo E. Inhibitors of serine/threonineprotein phosphatases antagonize the antinociception induced by agonists ofa2 adrenoceptors and GABAB but not k-opioid receptors in the tail flick test inmice. Pain 2005;114:212–20.

[25] Haley TJ, McCormick WG. Pharmacological effects produced by intracerebralinjection of drugs in the conscious mouse. Br J Pharmacol Chemother1957;12:12–5.

[26] Ocana M, Barrios M, Baeyens JM. Cromakalim differentially enhances anti-nociception induced by agonists of alpha2 adrenoceptors, g-aminobutyricacidB, mu and kappa opioid receptors. J Pharmacol Exp Ther 1996;276:1136–42.

[27] Therien AG, Blostein R. Mechanisms of sodium pump regulation. Am J PhysiolCell Physiol 2000;279:C541–66.

[28] Blanco G, Mercer RW. Regulation of the a2b1 and a3b1 isozymes of the Na+,K+-ATPase by Ca2+, PKA, and PKC. Ann N Y Acad Sci 1997;834:572–5.

[29] Fisone G, Snyder GL, Aperia A, Greengard P. Na+,K+-ATPase phosphorylation inthe choroid plexus: synergistic regulation by serotonin/protein kinase C andisoproterenol/cAMP-PK/PP-1 pathways. Mol Med 1998;4:258–65.

[30] Froehlich JP, Taniguchi K, Fendler K, Mahaney JE, Thomas DD, Albers RW.Complex kinetic behavior in the Na+,K+- and Ca2+-ATPases. Ann N Y Acad Sci1997;834:280–96.

[31] Lingrel JB, Kuntzweiler T. Na+,K+-ATPase. J Biol Chem 1994;269:19659–62.[32] Viola MS, Rodriguez de Lores Arnaiz G. Brain Na+,K+-ATPase isoforms: different

hypothalamus and mesencephalon response to acute desipramine treatment.Life Sci 2007;81:228–33.

[33] Acuna Castroviejo D, del Aguila CM, Fernandez B, Gomar MD, Castillo JL.Characterization of ouabain high-affinity binding to rat cerebral cortex. Mod-ulation by melatonin. Eur J Pharmacol 1992;226:59–67.

[34] Wu ZQ, Chen J, Chi ZQ, Liu JG. Involvement of dopamine system in regulation ofNa+,K+-ATPase in the striatum upon activation of opioid receptors by mor-phine. Mol Pharmacol 2007;71:519–30.

[35] Li P, Maguma HT, Thayne K, Davis B, Taylor DA. Correlation of the time courseof development and decay of tolerance to morphine with alterations in sodiumpump protein isoform abundance. Biochem Pharmacol 2010;79:1015–24.

[36] Andersson RM, Cheng SXJ, Aperia A. Forskolin-induced down-regulation ofNa+,K+-ATPase activity is not associated with internalization of the enzyme.Acta Physiol Scand 1998;164:39–46.

[37] Nestler EJ, Aghajanian GK. Molecular and cellular basis of addiction. Science1997;278:58–63.

[38] Gabra BH, Bailey CP, Kelly E, Smith FL, Henderson G, Dewey WL. Pre-treatmentwith a PKC or PKA inhibitor prevents the development of morphine tolerancebut not physical dependence in mice. Brain Res 2008;1217:70–7.

[39] Fisone G, Cheng SX-J, Nairn AC, Czernik AJ, Hemmings HC, Hoog JO, et al.Identification of the phosphorylation site for cAMP-dependent protein kinaseon Na+,K+-ATPase and effects of site-directed mutagenesis. J Biol Chem1994;269:9368–73.

[40] Logvinenko NS, Dulubova I, Fedosova N, Larsson SH, Nairn AC, Esmann M, et al.Phosphorylation by protein kinase C of serine-23 of the a1-subunit of ratNa+,K+-ATPase affects its conformational equilibrium. Proc Natl Acad Sci USA1996;93:9132–7.