Wh

ASNa

b

c

d

e

f

g

a

ARR1A

KWMAHB

I

acppo2c

(

0h

Phytomedicine 21 (2014) 745–752

Contents lists available at ScienceDirect

Phytomedicine

jou rn al homepage: www.elsev ier .de /phymed

ithania somnifera root extract prolongs analgesia and suppressesyperalgesia in mice treated with morphine

lessandro Orrùa,∗, Giorgio Marchesea, Gianluca Casua, Maria Antonietta Casua,anjay Kastureb, Filippo Cottigliac, Elio Acquasd,e,f, Maria Paola Masciag,icola Anzanic, Stefania Ruiua,∗

CNR-Institute of Translational Pharmacology, U.O.S. of Cagliari, Science and Technology Park of Sardinia Polaris, Pula, ItalySanjivani College of Pharmaceutical Education & Research, Kopargaon, IndiaDepartment of Life and Environmental Sciences – Drug Sciences Section, University of Cagliari, ItalyDepartment of Life and Environmental Sciences – Pharmaceutical, Pharmacological and Nutraceutical Sciences Section, University of Cagliari, ItalyCentre of Excellence on Neurobiology of Addiction, University of Cagliari, ItalyNational Institute of Neuroscience – INN, University of Cagliari, ItalyCNR-Institute of Neuroscience, Cagliari, Cittadella Universitaria, 09042 Monserrato, Italy

r t i c l e i n f o

rticle history:eceived 23 August 2013eceived in revised form6 September 2013ccepted 17 October 2013

eywords:ithania somniferaorphine

ntinociceptionyperalgesiainding assay

a b s t r a c t

Previous studies demonstrated that Withania somnifera Dunal (WS), a safe medicinal plant, prevents thedevelopment of tolerance to the analgesic effect of morphine.

In the present study, we investigated whether WS extract (WSE) (100 mg/kg, i.p.) may also modulatethe analgesic effect induced by acute morphine administration (2.5, 5, 10 mg/kg, s.c.) in the tail-flick andin the hot plate tests, and if it may prevent the development of 2.5 mg/kg morphine-induced reboundhyperalgesia in the low intensity tail-flick test. Further, to characterize the receptor(s) involved in theseeffects, we studied, by receptor-binding assay, the affinity of WSE for opioid (�, ı, k), cannabinoid (CB1,CB2), glutamatergic (NMDA), GABAergic (GABAA, GABAB), serotoninergic (5HT2A) and adrenergic (�2)receptors.

The results demonstrated that (i) WSE alone failed to alter basal nociceptive threshold in both tests,(ii) WSE pre-treatment significantly protracted the antinociceptive effect induced by 5 and 10 mg/kg ofmorphine only in tail-flick test, (iii) WSE pre-treatment prevented morphine-induced hyperalgesia in the

low intensity tail-flick test, and (iv) WSE exhibited a high affinity for the GABAA and moderate affinityfor GABAB, NMDA and ı opioid receptors.

WSE prolongs morphine-induced analgesia and suppresses the development of morphine-inducedrebound hyperalgesia probably through involvement of GABAA, GABAB, NMDA and ı opioid receptors.This study suggests the therapeutic potential of WSE as a valuable adjuvant agent in opioid-sparing

therapies.

ntroduction

Pain has negative consequences on health status and life quality,nd its relief is considered one of the major and complex medicaloncerns. To date, opioid analgesics remain the most efficaciousharmacological agents for the treatment of moderate to severeain, but their therapeutic benefit is often hampered by the devel-

pment of analgesic tolerance and hyperalgesia (Benyamin et al.008). Opioid-induced hyperalgesia (OIH), defined as a paradoxi-al increase in sensitivity to noxious stimuli (Pasero and McCaffery

∗ Corresponding authors. Tel.: +39 070 9241057; fax: +39 070 9241057.E-mail addresses: alessandro.orru@ift.cnr.it (A. Orrù), stefania.ruiu@cnr.it

S. Ruiu).

944-7113/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.phymed.2013.10.021

© 2013 Elsevier GmbH. All rights reserved.

2012), is a side effect difficult to recognize because its clinicalmanifestations are similar to those observed in opioid-inducedanalgesic tolerance (DuPen et al. 2007; Raffa and Pergolizzi 2011).Despite such difficulty, OIH incidence is growing as opioid pre-scriptions are increasing (Raffa and Pergolizzi 2011). A considerableeffort has been then expended in the development of novel ther-apeutic strategies that may maintain adequate opioid-inducedanalgesia and, at the same time, mitigate the appearance of tol-erance and OIH. Adjuvant pharmacological therapies are amongthe therapeutic approaches used to reach this purpose, and consistin the association of opioid analgesics with a second non-opioid

agent (DuPen et al. 2007; Khan et al. 2011). This approach hasbeen found to improve opioid-induced analgesia at different lev-els: enhancing pain relief (with a concurrent reduction of opioiddosage) and preventing, at the same time, the development of

7 edicin

smmaa

cooafidiapNma

oOccgpcspe

bmcTiatbtesif(gr

M

S

fho(wt

aw(GR

46 A. Orrù et al. / Phytom

ide effects, specifically tolerance and OIH (DuPen et al. 2007). Theain adjuvant agents that are currently associated to opioids in theanagement of pain are NMDA receptor antagonists, �2 receptor

gonists and cyclooxigenase-2 inhibitors (Low et al. 2012; Paserond McCaffery 2012).

Medicinal plants are among the potential adjuvant agents thatould be also evaluated in this therapeutic approach. Phytotherapyffers, in fact, a valid support to conventional medicine in a numberf diseases (Patwardhan 2005) and this also applies to the man-gement of pain (Low Dog 2008). Withania somnifera Dunal (WS,amily: Solanaceae) also known as Ashwagandha or Winter Cherrys a plant commonly used in Ayurvedic medicine to treat severaliseases (Alam et al. 2012). A growing number of pharmacolog-

cal studies demonstrated that WS extract (WSE) has anticancer,nti-inflammatory, immunomodulatory, adaptogenic and neuro-rotective properties (Alam et al. 2012). Interestingly, Kulkarni andinan (1997) demonstrated that, in mice chronically treated withorphine, WSE prevented the development of tolerance to the

ntinociceptive effect of morphine.The molecular underpinnings responsible for the development

f analgesic tolerance show similarities with those involved inIH (DuPen et al. 2007; Raffa and Pergolizzi 2011), and are theonsequence of plastic changes in neurotransmitter systems thatontribute to mediate pain transmission and opioid-induced anal-esia. Accordingly, pre-clinical studies demonstrated that severalharmacological agents prevent the development both of antinoci-eptive tolerance and OIH (Bryant et al. 2006; Gupta et al. 2011),uch as NMDA antagonists, among others. Moreover, the same com-ounds have been found to prolong or increase the acute analgesicffect of opioids (Fischer et al. 2005).

Starting from these studies, since the administration of WSElocked the development of tolerance to the analgesic effect oforphine (Kulkarni and Ninan 1997), we hypothesize that WSE

ould modulate additional aspects of morphine-induced analgesia.herefore, we aimed to determine whether the combined admin-stration of WSE with morphine could enhance or prolong thentinociceptive effect of morphine in the tail-flick and hot plateests. Moreover, we investigated the potential ability of WSE tolock the development of morphine-induced hyperalgesia usinghe low intensity tail-flick test, a validated model of OIH (Guptat al. 2011). Finally, since little information exists regarding theites of action of WSE in the brain and about the receptor(s)nvolved in its pharmacological and behavioural properties, per-orming competition studies we determined WSE affinity for opioid�, ı, k), cannabinoid (CB1, CB2), glutamatergic (NMDA), GABAer-ic (GABAA, GABAB), serotoninergic (5HT2A) and adrenergic (�2)eceptors.

aterials and methods

ubjects

Male CD1 mice (Charles River, Calco, Italy) 20–25 g were usedor both behavioural and receptor binding studies. Animals wereoused in an animal facility on a 12 h light/dark cycle (lightsn from 07:00 AM), at a constant room temperature of 21 ± 1 ◦Crelative humidity approximately 60%). Standard rodent chow andater were available ad libitum. Animals were allowed to adapt to

he animal facility conditions for at least two weeks after arrival.Procedures involving animals and their care were conducted in

ccordance with the institutional guidelines that are in compliance

ith national (D.L. 116/1992) and international laws and policies

EEC Council Directive 86/609, OJL 358, 1, December 12, 1987;uide for the Care and Use of Laboratory Animals, U.S. Nationalesearch Council, 1996). Every effort was made to minimize animal

e 21 (2014) 745–752

pain and discomfort and to reduce the number of experimentalsubjects.

Drugs and chemicals

Morphine hydrochloride (Salars, Como, Italy) was dissolved insaline (NaCl 0.9%) and administered subcutaneously (s.c.) in a vol-ume of 5 ml/kg. The standardized root methanolic extract of WS(Withania somnifera Dunal, Natural Remedies Pvt. Ltd., Bangalore,India) was dissolved in saline for analgesia experiments [adminis-tered intraperitoneally (i.p.) in a volume of 5 ml/kg] and in dimethylsulfoxide (DMSO, Sigma–Aldrich, Milan, Italy) for binding assay.The dose of WSE for analgesia experiments was selected on thebasis of previous studies (Kasture et al. 2009; Ruiu et al. 2013). [3H]-DAMGO ([d-Ala2, N-Me-Phe4,Gly-ol5]-enkephalin), [3H]-DPDPE([D-Pen2, D-Pen5]-enkephalin), [3H]-U-69,593, [3H]-CP55,950,[3H]-MK801, [3H]-Muscimol, [3H]-Baclofen, [3H]-Clonidine and[3H]-Ketanserine were purchased from Perkin Elmer, Monza (MB),Italy. CP-55,940, Naloxone, U-69,593, muscimol, MK801, yohim-bine and methysergide were obtained from Tocris Cookson Ltd.(Bristol, UK).

Plant materials

The standardized WSE was kindly provided by Natural RemediesPvt. Ltd., Bangalore, India.

High-performance liquid chromatographic (HPLC) analysis

The WSE has been characterized by an HPLC-fingerprint analy-sis, as certified by Natural Remedies Pvt. Ltd., with identification ofthe main withanolides. This analysis with the necessary descriptionof the technique has been published in Neurotoxicity Research byKasture et al. (2009). An HPLC system (Shimadzu, LC 2010 A, Japan)equipped with UV detector, auto-injector, and column oven withclass VP software was used. The stationary phase was an octadecyl-silane column (Phenomenex-Luna, C18, 5 �m, 250 mm × 4.6 mm).The mobile phase was a mixture of phosphate buffer (SolventA) [prepared by dissolving 0.136 g of KH2PO4 in 900 ml of HPLCgrade water and by adding 10% dilute aqueous H3PO4, adjus-ting the pH to 2.8 ± 0.05 and making the volume of 1000 mlwith water] and acetonitrile (Solvent B). The following with-anolides were identified: withanoside-IV, 0.49%w/w; physagulinD, 0.11%w/w; 27-hydroxywithanone, 0.01%w/w; withanoside-V,0.33%w/w; withaferin-A, 0.11%w/w; 12-deoxy withastramonolide,0.16%w/w; withanolide-A, 0.19%w/w; withanone 0.004%w/w, andwithanolide-B, 0.03%w/w.

Analgesia experiments

The antinociceptive effects were quantified using the tail-flicktest and the hot plate test (Ruiu et al. 2003). An automated device(model 7360, Ugo Basile, Italy) was used to determine the tail-flicklatency, defined by the time (s) at which the animals withdraw thetail from a radiant heat source. Mice were held and gently restrainedabove the apparatus; the light beam was focused 1.5 cm from thetip of the ventral surface of the tail. The stimulus intensity wasadjusted to result in a mean pre-drug control latency of 2–3 s, anda cut-off time of 12 s was applied to avoid tissue damage.

A semi-automated device (model 7280, Ugo Basile, Italy) wasused to determine the reaction of mice placed on the hot plate,defined by the time (s) at which mice exhibited a nociceptive

response or discomfort (licking or fanning the paws, jumping). A50 cm high Plexiglas cylinder was suspended over the hot plate andthe temperature was maintained at 55 ± 0.2 ◦C; to avoid skin dam-age, after 15 s mice were removed from the hot plate. Mice were

edicin

pm5Wu

M

ittmw3a(3

S

pmotcbchtla

R

aRhsAwwinr

wat5b[fw

isnswaprT

A. Orrù et al. / Phytom

re-treated with saline or WSE 100 mg/kg 30 min before saline ororphine treatment; increasing doses of morphine were used (2.5,

and 10 mg/kg). Basal algesia was assessed right before saline orSE pre-treatment (baseline). The effects of the drugs were eval-

ated 30, 60, 120, 240 and 360 min after morphine treatment.

orphine-induced hyperalgesia experiment

The ability of WSE to inhibit the development of morphine-nduced hyperalgesia was evaluated using the low-intensityail-flick test according Gupta et al. (2011) with few modifica-ions. Briefly, the stimulus intensity was adjusted to result in a

ean pre-drug control latency of 5–7 s, and a cut-off time of 10 sas applied. Mice were pre-treated with saline or WSE 100 mg/kg

0 min before saline or morphine treatment (2.5 mg/kg). Basallgesia was assessed right before saline or WSE pre-treatmentbaseline). The effects of the drugs were evaluated 30, 60, 120, 240,00, 360 and 420 min after morphine treatment.

pontaneous and morphine-induced motor activity experiments

In order to evaluate the ability of WSE to modulate furtherharmacological effects induced by morphine, spontaneous andorphine-induced motor activity experiments have been carried

ut (Ruiu et al. 2013). Mice were tested in chambers made ofransparent Plexiglas (40L cm × 30W cm × 40H cm) interfaced to aomputer equipped with a TSE software (TSE Systems, Bad Hom-urg, Germany). The distance travelled (m) by mice inside thehambers was detected by 8 horizontal photocells. Mice wereabituated to the chambers for 30 min. Immediately after habitua-ion, mice were treated with saline or WSE 100 mg/Kg and 30 minater were injected with saline or morphine (5 or 10 mg/Kg). Motorctivity was evaluated for 360 min after morphine administration.

eceptor binding experiments

[3H]-DAMGO – [3H]-DPDPE – [3H]-U69,593. Ligand bindingssays were carried out according to the procedure described byuiu et al. (2013). Briefly, the whole brain minus cerebellum wasomogenized in Tris–HCl (pH 7.4), centrifuged at 48,000 × g, re-uspended in the same buffer solution, and incubated at 37 ◦C.fter a further centrifugation step at 48,000 × g, the final pelletas re-suspended in the same buffer solution. Brain membranesere incubated with [3H]-DAMGO, [3H]-DPDPE or [3H]-U69,593

n Tris–HCl buffer at 25 ◦C for 60 min in the absence or presence ofaloxone (1 �M) (for � and ı receptors) or U 69,593 (10 �M) (for keceptors).

[3H]-CP55,940. The whole brain minus cerebellum and spleenere homogenized in TME buffer (50 mM Tris–HCl, 1 mM EDTA

nd 3.0 mM MgCl2, pH 7.4), centrifuged at 1086 × g for 10 min andhe resulting supernatants were centrifuged at 45,000 × g. [3H]-CP-5,940 binding was performed by the method previously describedy Ruiu et al. (2003). Briefly, the membranes were incubated with

3H]-CP55,940 for 1 h at 30 ◦C in TME buffer containing 5 mg/ml ofatty acid-free bovine serum albumin (BSA). Non-specific bindingas estimated in the presence of 10 �M of CP55,940.

[3H]-Muscimol. Ligand binding assays were carried out accord-ng to the procedure described by Beaumont et al. (1978) withlight modifications. Briefly, the cerebral cortex were homoge-ized in 0.32 M sucrose, centrifuged at 1000 × g and the resultingupernatants were centrifuged at 20,000 × g. The resulting pelletas suspended in ice-cold water, homogenized and centrifuged

t 8000 × g. The supernatant together with the buffy layer on theellet was then centrifuged at 48,000 × g. The resulting pellet wase-suspended in water, and once more centrifuged at 48,000 × g.he final pellet was frozen and stored at −80 ◦C. On the day of

e 21 (2014) 745–752 747

analysis, membrane pellet was allowed to thaw at 4 ◦C before re-suspension in 50 mM Tris–citrate buffer, pH 7.1, containing 0.05%Triton X-100 and incubated for 30 min at 37 ◦C. Following incuba-tion, the suspension was centrifuged at 48,000 × g. The washingstep was repeated three more times and the final pellet was thenre-suspended in the binding buffer. Non-specific binding was esti-mated in the presence of 200 �M of GABA.

[3H]-MK801. Ligand binding assays were carried out accordingto the procedure described by Wong et al. (1986) with slight mod-ifications. Briefly, the cerebral cortex were homogenized in 0.32 Msucrose, centrifuged at 1000 × g for 10 min at 4 ◦C and the resultingsupernatants were centrifuged at 20,000 × g. The resulting pelletwas re-suspended in ice-cold water and centrifuged at 8000 × g.The supernatant together with the buffy layer on the pellet was thencentrifuged at 48,000 × g. The resulting pellet was re-suspended in5 mM Tris–HCl buffer (pH 7.4) containing 5 mM EDTA and incu-bated at 37 ◦C for 30 min prior to centrifugation at 48,000 × g. Thewashing step was repeated once more times and the final pelletwas frozen and stored at −80 ◦C. On the day of analysis, mem-brane pellet was allowed to thaw at 4 ◦C before re-suspension in5 mM Tris–HCl buffer and incubated at 37 ◦C. Following incubation,the suspension was centrifuged at 48,000 × g. The resulting pelletwas re-suspended in ice-cold water, and once more centrifuged at48,000 × g. The washing step was repeated once more times andthe final pellet was then re-suspended in 5 mM Tris–HCl buffer.Nonspecific binding was defined by 100 �M unlabeled MK-801.

[3H]-Clonidine. Ligand binding assays were carried out accord-ing to the procedure described by Barturen and Garcia-Sevilla(1992) with slight modifications. Briefly, the cerebral cortex werehomogenized in Tris–HCl containing 0.32 M sucrose, centrifugedat 1000 × g for 10 min and the resulting supernatants were cen-trifuged at 31,000 × g. Cerebral cortex membranes were incubatedwith [3H]-Clonidine in Tris–HCl buffer containing 10 mM MgCl2and 250 �M ascorbic acid at 25 ◦C for 60 min in the absence orpresence of yohimbine (50 �M).

[3H]-Baclofen.Ligand binding assays were carried out accordingto the procedure described by Ruiu et al. (2013). Briefly, the wholebrain minus cerebellum was homogenized in 0.32 M sucrose con-taining 1 mM EDTA. The homogenate was centrifuged at 10,000 × gand the supernatant collected and re-centrifuged at 20,000 × g. Thepellet was re-suspended in ice-cold water, homogenized and cen-trifuged at 8000 × g. The supernatant together with the buffy layeron the pellet was then centrifuged at 45,000 × g. The resultingpellet was re-suspended in ice-cold water and once more cen-trifuged at 45,000 × g. The final pellet was frozen and stored at−80 ◦C. To this end, membrane pellets were allowed to thaw at4 ◦C before re-suspension in 50 mM Tris–HCl. The suspension wasincubated at 20 ◦C before centrifugation at 7000 × g. The washingstep was repeated three more times allowing 15 min incubationwith each addition of the same buffer to remove the endogenousligand GABA. The final pellet was then re-suspended in appropri-ate binding buffer. The incubation buffer was: 50 mM Tris–HCl (pH7.4) containing 2.5 mM CaCl2. GABA (100 �M) was used to definenon-specific binding.

[3H]-Ketanserin. The assays for [3H]-ketanserin binding wereperformed as previously described (Gauffre et al. 1997) with slightmodifications. Briefly, the cerebral cortex were homogenized in50 mM Tris–HCl buffer and centrifuged at 48,000 × g. Pellets werewashed twice by resuspension in the same buffer followed bycentrifugation at 48,000 × g. The final pellets were then stored at−80 ◦C. When needed, pellets were suspended in Tris–HCl bufferand the membranes were incubated in the presence of [3H]-

ketanserin at 37 ◦C. Non-specific binding was assessed by adding10 �M of methysergide.

Analysis of samples. Displacement curves were carried outusing serial dilutions ranging from 2 mg/ml to 0.001 mg/ml of the

7 edicine 21 (2014) 745–752

uletBr2G

D

rb[(tflmhs(sm

etm(bt

p

nPvvtw

R

E

itcdettab

oaFs5poPv

Fig. 1. Effect of WSE pre-treatment on morphine-induced analgesia in the tail-flick test. Saline or WSE (100 mg/kg, i.p.) were administered 30 min before 2.5 (A),5 (B) and 10 (C) mg/kg, s.c. morphine injection. Basal algesia was assessed rightbefore saline or WSE pre-treatment (baseline). The effects of the drugs were eval-uated 30, 60, 120, 240 and 360 min after morphine treatment. (See Materials andMethods for further details). Results are expressed as %MPE. Each point representsthe mean ± S.E.M. of 6–10 mice per group. *p < 0.05 vs. saline + saline-treated mice;

48 A. Orrù et al. / Phytom

nlabelled compound. To avoid possible undesired effects on radio-igand binding, DMSO concentration in the different assays neverxceeded 1% (v/v). The binding reaction were stopped by rapid fil-ration under vacuum through glass-fibre filters (Whatman) using arandell 36-sample harvester (Gaithesburg, MD, USA). Filter-boundadioactivity was counted in a liquid scintillation counter (Tricarb900; PerkinElmer Life Sciences, Boston, MA, USA) using Ultimaold (Packard, USA) as scintillation fluid.

ata analysis

Treatment-induced variations in tail-flick and hot plateesponse were calculated as the percentage of maximal possi-le effect (MPE) according to the following formula: MPE [%]:(T1-T0)/(T2-T0)] × 100, where T0 and T1 are the latency beforebaseline) and after treatment, and T2 is the cut-off time (12 s inhe tail-flick, 15 s in the hot plate and 10 s in the low-intensity tail-ick tests). Data from behavioural experiments are expressed asean ± standard error (S.E.M.) of %MPE. The %MPE in tail-flick and

ot plate experiments were analyzed separately by repeated mea-ure three-way analysis of variance (ANOVA) with pre-treatmentsaline and WSE) and treatment (saline and morphine) as between-ubjects factors, and time as within-subjects factor (repeatedeasures).In the spontaneous and morphine-stimulated motor activity

xperiments, the average distance travelled (m) after WSE pre-reatment and morphine treatment was analyzed by repeated

easures three-way ANOVA, with pre-treatment (pre-treat)saline and WSE) and treatment (treat) (saline and morphine) asetween-subjects factors and time (t) (min) as within-subjects fac-or (repeated measures).

For each of the statistical analysis described above, when appro-riate, post hoc comparisons were done using Tukey’s test.

Data from radioligand inhibition experiments were analyzed byonlinear regression analysis of a Sigmoid Curve using GraphPadrism program (Graph Pad Software, Inc., San Diego, CA, USA). IC50alues were derived from the calculated curves and converted to Kialues as described previously (Cheng and Prusoff 1973). All recep-or binding experiments were performed in triplicate and resultsere confirmed in at least four independent experiments.

esults

ffects of WSE on morphine-induced analgesia

The effects of WSE (100 mg/kg) pre-treatment on morphine-nduced analgesia were evaluated using the tail-flick (Fig. 1) andhe hot plate tests (Fig. 2). WSE alone failed to alter the noci-eptive reaction time. Morphine (2.5, 5 and 10 mg/kg) elicited aose-dependent antinociceptive effect in both tests, although itsfficacy was higher when the tail-flick test was used. The peak inhe analgesic effect was reached 60 min after morphine adminis-ration. Co-administration of WSE with morphine modulated itsnalgesic activity depending on the dose of morphine and on theehavioural test used.

In the tail-flick test (Fig. 1) morphine alone at the dosef 2.5 mg/kg increased the tail-flick latency along 60 minfter its administration [Ftreat(1,30) = 46.16, p < 0.0001;treat × t(4,120) = 8.35, p < 0.0001; Tukey’s test, p < 0.05 vs.aline + saline-treated mice], and along 120 min following

mg/kg [Ftreat(1,26) = 135.99, p < 0.0001; Ftreat × t(4,104) = 10.97,

< 0.0001; Tukey’s test, p < 0.05 vs. saline + saline-treated mice]r 10 mg/kg morphine administration [Ftreat(1,31) = 260.48,

< 0.0001; Ftreat × t(4,124) = 5.61, p < 0.005; Tukey’s test, p < 0.05s. saline + saline-treated mice]. Interestingly, although WSE

#p < 0.05 vs. saline + morphine-treated mice (three-way ANOVA followed by Tukey’spost hoc test).

co-administration failed to modulate the antinociceptiveeffect elicited by 2.5 mg/kg morphine [Fpre-treat(1,30) = 0.78,not significant (N.S.); Fpre-treat × treat(1,30) = 2.28, N.S.;Fpre-treat × t(4,120) = 0.66, N.S.; Fpre-treat × treat × t(4,120) = 1.19, N.S.],it protracted the analgesic effect induced by 5 mg/kg morphine[Fpre-treat(1,26) = 5.76, p < 0.05; Fpre-treat × treat(1,26) = 7.66, p < 0.05;Fpre-treat × t(4,104) = 2.82, p < 0.05; Fpre-treat × treat × t(4,104) = 2.66,p < 0.05], and 10 mg/kg morphine [Fpre-treat(1,31) = 2.10, N.S.;Fpre-treat × treat(1,31) = 3.57, N.S.; Fpre-treat × t(4,124) = 19.54,p < 0.0001; Fpre-treat × treat × t(4,124) = 2.85, p < 0.05). In fact, theanalgesic effect was still evident 240 min after 5 mg/kg morphine

treatment in mice pre-treated with WSE (78 ± 11%MPE) com-pared to mice pre-treated with saline (16 ± 7%MPE) (Tukey’s test,p < 0.05 vs. saline + saline-treated mice and vs. saline + morphine-treated mice). Likewise, analgesia was still evident at 240 min

A. Orrù et al. / Phytomedicine 21 (2014) 745–752 749

Fig. 2. Effect of WSE pre-treatment on morphine-induced analgesia in the hot platetest. Saline or WSE (100 mg/kg, i.p.) were administered 30 min before 2.5 (A), 5 (B)and 10 (C) mg/kg, s.c. morphine. Basal algesia was assessed right before saline orWSE pre-treatment (baseline). The effects of the drugs were evaluated 30, 60, 120,240 and 360 min after morphine treatment. (See Materials and Methods for furtherd6f

(pm2m

(apt[TaFst[F

Fig. 3. Effect of WSE pre-treatment on morphine-induced hyperalgesia in the low-intensity tail-flick test. Saline or WSE (100 mg/kg, i.p.) were administered 30 minbefore morphine (2.5 mg/kg, s.c.). Basal algesia was assessed right before salineor WSE pre-treatment (baseline). WSE pre-treatment (baseline). The effects of thedrugs were evaluated 30, 60, 120, 240, 300, 360 and 420 min after morphine treat-ment. (See Materials and Methods for further details). Results are expressed as

etails). Results are expressed as %MPE. Each point represents the mean ± S.E.M. of–10 mice per group. *p < 0.05 vs. saline + saline-treated mice (three-way ANOVAollowed by Tukey’s post hoc test).

55 ± 7%MPE) and 360 min (45 ± 11%MPE) after 10 mg/kg mor-hine treatment, in mice pre-treated with WSE when compared toice pre-treated with saline, in which the %MPE was respectively

0 ± 4% and 10 ± 3% (Tukey’s test, p < 0.05 vs. saline + saline-treatedice and vs. saline + morphine-treated mice).A different picture was observed in the hot plate test

Fig. 2) in which morphine, at the dose of 2.5 mg/kg had nonalgesic effect [Ftreat(1,30) = 0.35, N.S.; Ftreat × t(4,120) = 5.4,

< 0.0001]; on the contrary, 5 and 10 mg/kg increasedhe reaction time respectively along the first 60 minFtreat(1,26) = 46.53, p < 0.0001; Ftreat × t(4,104) = 5.61, p < 0.0001;ukey’s test, p < 0.05 vs. saline + saline-treated mice], andlong the first 120 min [Ftreat(1,31) = 58.72, P < 0.0001;treat × t(4,124) = 10.27, p < 0.0001; Tukey’s test, p < 0.05 vs.

aline + saline-treated mice]. WSE pre-treatment failedo modulate the analgesic effect induced by 2.5 mg/kgFpre-treat(1,30) = 0.41, N.S.; Fpre-treat × treat(1,30) = 0.17, N.S.;pre-treat × t(4,120) = 0.17, N.S.; Fpre-treat × treat × t(4,120) = 0.28, N.S.],

%MPE. Each point represents the mean ± S.E.M. of 7–11 mice per group. *p < 0.05 vs.saline + saline-treated mice; #p < 0.05 vs. saline + morphine-treated mice (three-wayANOVA followed by Tukey’s post hoc test).

5 mg/kg [Fpre-treat(1,26) = 0.0003, N.S.; Fpre-treat × treat(1,26) = 0.27,N.S.; Fpre-treat × t(4,104) = 0.82, N.S.; Fpre-treat × treat × t(4,104) = 0.48,N.S.] and 10 mg/kg morphine [Fpre-treat(1,31) = 0.07, N.S.;Fpre-treat × treat(1,31) = 0.30, N.S.; Fpre-treat × t(4,124) = 1.18, N.S.;Fpre-treat × treat × t(4,124) = 0.55, N.S.].

Effect of WSE on morphine-induced hyperalgesia

The effects induced by WSE (100 mg/kg) pre-treatment onmorphine-induced hyperalgesia were evaluated using the low-intensity tail-flick test (Fig. 3). WSE alone failed to alter thenociceptive reaction time. In agreement with Gupta et al. (2011),we found that a single injection of a low dose of morphine(2.5 mg/kg) elicited a biphasic effect in the nociceptive response ofCD1 mice. In fact, analgesia was reached at 30 and 60 after morphineadministration [Ftreat(1,32) = 13.91, p < 0.001. Ftreat × t(6,192) = 2.98,p < 0.01; Tukey’s test, p < 0.05 vs. saline + saline-treated mice]. Thiseffect was followed by a gradual appearance of hyperalgesia;Tukey’s test revealed that morphine administration reduced thenociceptive response (−65 ± 13%MPE) when compared to the noci-ceptive response of control mice (−4 ± 6%MPE) 360 min aftermorphine administration (p < 0.05). Consistently with the resultsshown in Fig. 1A, pre-treatment with WSE did not potentiatethe analgesic effect of morphine during the first 120 min, how-ever, it significantly prevented morphine-induced hyperalgesia[Fpre-treat(1,32) = 7.25, p < 0.05; Fpre-treat × treat(1,32) = 9.40, p < 0.005;Fpre-treat × t(6,192) = 2.98, p < 0.001; Fpre-treat × treat × t(6,192) = 1.67,N.S.]. Tukey’s test revealed that at 240, 300, 360 and 420 minafter morphine administration the nociceptive threshold of micepre-treated with WSE was higher when compared with the oneobserved in mice pre-treated with saline (p < 0.05).

Effect of WSE on morphine-induced hyper-locomotion

Fig. 4 shows the effects of WSE on spontaneous motor

activity and on morphine-induced (5 and 10 mg/kg) hyper-locomotion. Repeated measures three-way ANOVA of theaverage distances travelled revealed that morphine increasedmotor activity in a dose dependent manner [Ftreat(2,38) = 18.43,

750 A. Orrù et al. / Phytomedicine 21 (2014) 745–752

Fig. 4. Effect of WSE pre-treatment on morphine-induced hyper-locomotion. After 30 min of habituation in the motor activity chambers, mice were pre-treated with salineo stratioa MethE saline

pvblNNmapd

C

iitgaatwNatr

D

ma

TK

r WSE (100 mg/kg, i.p.) 30 min before saline or morphine (10 mg/kg, s.c.) adminictivity chambers, and motor activity was recorded for 360 min. (See Materials andach bar represents the mean ± S.E.M. of 5–11 mice per group. *p < 0.05 vs. saline +

< 0.0001; Ftreat × t(8,152) = 19.61, p < 0.0001; Tukey’s test, p < 0.05s. saline + saline-treated mice]. WSE pre-treatment failed to alteroth spontaneous motor activity and morphine-induced hyper-

ocomotion [Fpre-treat(1,38) = 0.09, N.S.; Fpre-treat × treat(2,38) = 0.18,.S.; Fpre-treat × t(4,152) = 0.11, N.S.; Fpre-treat × treat × t(8,152) = 0.60,.S.]. Specifically, Tukey’s post hoc analysis revealed that 10 mg/kgorphine, alone and in combination with WSE increased motor

ctivity 120, 240 and 360 min after administration (Tukey’s test, < 0.05 vs. saline + saline-treated mice), but WSE pre-treatmentid not affect morphine-induced hyper-locomotion.

haracterization of WSE binding properties

We evaluated whether the behavioural effects of WSE observedn the present study could be associated to the receptors involvedn pain regulation. Therefore, we examined the affinity of WSEowards opioid (�, ı, k), cannabinoid (CB1, CB2), glutamater-ic (NMDA), GABAergic (GABAA, GABAB), serotoninergic (5HT2A)nd adrenergic (�2) receptors using radioligand receptor bindingssays. As shown in Table 1, WSE exhibits the highest affinity forhe GABAA receptors. It is worth noting that WSE binds to GABAAith an affinity 10, 13 and 15 times greater than that for GABAB,MDA and ı opioid receptors respectively. In contrast, WSE showed

weak binding affinity to � and k opioid receptors and to CB1 recep-or and no binding affinity for the CB2, 5HT2A and �2 adrenergiceceptors.

iscussion

Our results demonstrate that the administration of WSE, aethanolic root extract of Withania somnifera, prolongs the acute

nalgesic effect of morphine in the tail-flick test and prevents

able 1i values for WSE were determined as described in details in the Methods section. Result

Ki (�g/ml)

� ı k CB1 CB2

385 ± 14 166 ± 11 775 ± 56 837 ± 74 >1000

n. Immediately after morphine administration, mice were returned to the motorods for further details). Results are expressed as average of distance travelled (m).

-treated mice (three-way ANOVA followed by Tukey’s post hoc test).

the development of hyperalgesia in a model of OIH. In addition,receptor binding assays suggest a possible involvement of GABAA,GABAB, NMDA and ı opioid receptors in the observed behaviouraleffects.

WSE was devoid of analgesic activity, in both the tail-flick andthe hot plate tests, confirming previous results in which WSE(100 mg/kg) failed to increase the tail-flick latencies in mice 60 minafter oral administration (Kulkarni and Ninan 1997). WSE also failedto strengthen the analgesic effect of morphine in the tail-flick testbut, surprisingly, it prolonged the length of morphine action in adose dependent manner. Indeed, while the analgesic effect inducedby 5 and 10 mg/kg morphine alone declined 2 hours after mor-phine administration, it lasted respectively 4 and 6 hours whenmorphine was combined with WSE. Notably, this pharmacologicaleffect appeared to be specific for morphine-elicited analgesia, sinceWSE pre-treatment failed to modify the hyper-locomotor activityinduced by morphine. Interestingly, we found that WSE protractedthe analgesic effect induced by morphine in the tail-flick test butnot in the hot plate test. The tail-flick and hot plate tests are twobehavioural models of acute pain thought to differ in the levelof nociceptive information processing (Franklin and Abbott 1989).Accordingly, it was suggested that the response to the tail-flick testmay predominantly reflect a spinally integrated nociceptive reflex,whereas the complex painful response assessed by the hot platetest is thought to require the involvement of higher brain func-tions. Thus, based on the present results it may be hypothesizedthat spinal rather than supraspinal mechanisms may account forthe different effects induced by WSE in the hot plate and tail flick

tests when associate with morphine.

Several hypotheses can be made to explain the ability of WSEto prolong morphine-induced analgesia in the tail flick test. Onepossible interpretation is that WSE might have altered morphine

s are mean ± SEM of four independent experiments assayed in triplicate.

GABAA GABAB NMDA �2 5-HT2A

13 ± 2 130 ± 12 193 ± 21 >1000 >1000

edicin

pc(cb

timaia(aprtaacutoSeNop

tedaCmpt2

matbWiNd(ıraaGdipGno1t(atte

A. Orrù et al. / Phytom

harmacokinetics, as previously observed for L-type calciumhannel blockers (Shimizu et al. 2004). However, Kasture et al.2009) demonstrated that WSE administration induced statisti-ally insignificant decrease, rather than increase, in plasma andrain morphine levels.

This observation brings us to conclude that the prolonga-ion of morphine analgesia induced by WSE is pharmacodynamicn nature. An interesting hypothesis is that WSE might prolong

orphine-induced analgesia by preventing the development ofcute tolerance. Acute analgesic tolerance following opioids admin-stration has been found to develop rapidly both in humans (Viniknd Kissin 1998) and in rodents (Kissin et al. 1991). Kissin et al.1991) have shown in rats that, under continuous infusion orfter a single subcutaneous injection, the analgesic effect of mor-hine rapidly declined despite brain levels of morphine haveemained stable over time. The molecular mechanism underlyinghe development of analgesic tolerance following acute morphinedministration is similar to that involved in chronic analgesic toler-nce (Fairbanks and Wilcox 1997). Consistently, several classes ofompounds, such as, NMDA receptor antagonists and GABAA mod-lators, have been found to prevent the development of analgesicolerance following both acute (Kissin et al. 1997, 2000) and chronicpioid administration (Bryant et al. 2006; Tejwani et al. 1993).ince WSE administration has been found to prevent analgesic tol-rance following chronic morphine administration (Kulkarni andinan 1997), the prolongation of morphine acute analgesic effectbserved in the present study can be ascribed to a WSE-inducedrevention of acute tolerance development.

Moreover, several studies suggest the involvement of OIH inhe development of acute tolerance (Milne et al. 2013); more gen-rally, although opioid tolerance and OIH should be consideredistinct phenomena they partially share the same cellular mech-nism, pathogenesis and therapeutic approach (DuPen et al. 2007).onsistently, using the low-intensity tail-flick test, that allows toagnify the development of hyperalgesia induced by a low mor-

hine dose (Gupta et al. 2011), we found that WSE antagonizedhe progressive reduction on nociceptive threshold induced by.5 mg/kg morphine.

In the attempt to elucidate the mechanism(s) by which WSEodulates the effects induced by morphine, we studied the

ffinity of WSE for several receptors involved in pain transduc-ion signalling and opioids-induced analgesia. These experimentsrought us to restrict the search of potential molecular targets ofSE to GABAA, GABAB, ı opioid and NMDA receptors. WSE exhib-

ted a high affinity for the GABAA and moderate affinity for GABAB,MDA and ı opioid receptors, confirming and extending previousata showing that WSE interacts with GABAA and GABAB receptorsMehta et al. 1991; Ruiu et al. 2013). Interestingly, GABA, NMDA and

opioid receptors-mediated neurotransmission plays an importantole in pain transduction signalling and is involved in the appear-nce of opioid-induced analgesic tolerance and hyperalgesia (Davisnd Pasternack 2005; Enna and McCarson 2006). In particular,ABA transmission may have opposing effects on pain processing,epending upon its location within the central nervous system;

ts enhancement elicits analgesia at the spinal level whereas isronociceptive at supraspinal level (Davis and Pasternack 2005;iordano 2005). Systemic administration of GABAA and GABAB ago-ists, as well as benzodiazepines have been found to potentiatepioid-induced analgesia (Aley and Kulkarni 1989; Bergman et al.988); moreover, pre-treatment with benzodiazepines attenuatedhe development of tolerance following acute opioid treatmentKissin et al. 1997). Thus, the observation that WSE binds the GABAA

nd GABAB receptors, combined with the finding that WSE con-ains a GABA-mimetic component (Mehta et al. 1991), suggests thathe GABA system may be partially responsible for the behaviouralffects observed in the present study.

e 21 (2014) 745–752 751

Glutamate is the main excitatory neurotransmitter involved inthe transmission of nociceptive stimuli at spinal level (Giordano2005); moreover, NMDA receptor sensitization into spinal neu-rons has been found to play a pivotal role in the development ofopioid-induced analgesic tolerance and OIH (DuPen et al. 2007).Consistently, co-administration of NMDA receptor antagonists hasbeen found to potentiate opioid-induced analgesia (Fischer et al.2005) and to prevent analgesic tolerance and OIH both in animalsand in humans (DuPen et al. 2007). Moreover, Zhao and Joo (2008)found that the enhancement of NMDA transmission responsiblefor the development of analgesic tolerance and OIH is mediatedby activation of ı opioid receptors. In this regard, our observationthat WSE moderately binds to NMDA and ı opioid receptors maysuggest that they may contribute to prolong morphine analgesiaand to prevent morphine-induced hyperalgesia. The involvementof NMDA receptors could also explain why we have observed aprolongation of morphine analgesia in the tail-flick but not in thehot plate test; accordingly, Kozela et al. (2001) demonstrated thatlow-affinity NMDA receptor antagonists preferentially potentiatemorphine-induced analgesia recorded from the tail but not fromthe hind paw.

In conclusion, our results suggest that WSE could be a valu-able adjuvant in opioid-sparing therapies, which aim to minimizeopioid dosage while providing optimal analgesia and limiting theemergence of tolerance and hyperalgesia. The concurrent actionof WSE on GABAA, GABAB, NMDA and ı opioid receptors couldbe responsible for the behavioural effects observed in the presentstudy. Further studies are needed to evaluate the net contributionof each of these receptors, and this will be achieved by identifyingthe single phytochemical components of WSE and characterizingtheir intrinsic activity.

Source of funding

From Regione Autonoma della Sardegna (RAS) (CRP 26805-CUPB71J11001480002) to SR, EA, FC.

Conflicts of interest

The authors declare they have no conflicts of interest.

Acknowledgements

The authors wish to thank Dr. Amit Agarwal (Natural RemediesPvt. Ltd., Bangalore, India) for the generous gift of standardizedWithania somnifera root extract. The authors are grateful to Mrs.Gabriella Manca and Mr. Antonio Pilleri for animal care.

References

Alam, N., Hossain, M., Khalil, M.I., Moniruzzaman, M., Sulaiman, S.A., Gan, S.H., 2012.Recent advances in elucidating the biological properties of Withania somniferaand its potential role in health benefit. Phytochemistry Reviews 11, 97–112.

Aley, K.O., Kulkarni, S.K., 1989. GABAergic agents-induced antinociceptive effect inmice. Methods and Findings in Experimental and Clinical Pharmacology (10),597–601.

Barturen, F., Garcia-Sevilla, J.A., 1992. Long term treatment with desipramineincreases the turnover of alpha 2-adrenoceptors in the rat brain. MolecularPharmacology 42 (5), 846–855.

Beaumont, K., Chilton, W., Yamamura, H.I., Enna, S.J., 1978. Muscimol binding in ratbrain: association with synaptic GABA receptors. Brain Research 148, 153–162.

Benyamin, R., Trescot, A.M., Datta, S., Buenaventura, R., Adlaka, R., Sehgal, N., Glaser,S.E., Vallejo, R., 2008. Opioid complications and side effects. Pain Physician 11(Suppl. 2), S105–S120.

Bergman, S.A., Wynn, R.L., Williams, G., 1988. Diazepam enhances fentanyl and

diminishes meperidine antinociception. Anesthesia Progress 35 (5), 190–194.

Bryant, C.D., Eitan, S., Sinchak, K., Fanselow, M.S., Evans, C.J., 2006. NMDA recep-tor antagonism disrupts the development of morphine analgesic tolerance inmale, but not female C57BL/6J mice. American Journal of Physiology, Regulatory,Integrative, and Comparative Physiology 291 (2), R315–R326.

7 edicin

C

D

D

E

F

F

F

G

G

G

K

K

K

K

K

K

52 A. Orrù et al. / Phytom

heng, Y., Prusoff, W.H., 1973. Relationship between the inhibition constant (K1)and the concentration of inhibitor which causes 50 per cent inhibition (I50) ofan enzymatic reaction. Biochemical Pharmacology 22 (23), 3099–3108.

avis, M.P., Pasternack, G.W., 2005. Opioid Receptors and Opioid Pharmacody-namics. In: Davis, M., Glare, P., Hardy, J. (Eds.), Opioids in Cancer Pain. OxfordUniversity Press, Oxford, UK, pp. 11–41.

uPen, A., Shen, D., Ersek, M., 2007. Mechanisms of opioid-induced tolerance andhyperalgesia. Pain Management Nursing 8 (3), 113–121.

nna, S.J., McCarson, K.E., 2006. The role of GABA in the mediation and perceptionof pain. Advances in Pharmacology 54, 1–27.

airbanks, C.A., Wilcox, G.L., 1997. Acute tolerance to spinally administered mor-phine compares mechanistically with chronically induced morphine tolerance.Journal of Pharmacology and Experimental Therapeutics 282 (3), 1408–1417.

ischer, B.D., Carrigan, K.A., Dykstra, L.A., 2005. Effects of N-methyl-d-aspartatereceptor antagonists on acute morphine-induced and l-methadone-inducedantinociception in mice. Journal of Pain 6 (7), 425–433.

ranklin, K., Abbott, F., 1989. Techniques for assessing the effects of drugs on nocicep-tive responses. In: Boulton, M., Baker, G., Greenshaw, A. (Eds.), Neuromethods,Psychopharmacology, vol. 13. The Human Press, Clifton, NJ, pp. 145–215.

auffre, J.C., Aguerre, S., Mormède, P., Chaouloff, F., 1997. Cortical [3H]ketanserinbinding and 5-HT2A receptor-mediated inositol phosphate production in thespontaneously hypertensive rat and Lewis rat strains. Neuroscience Letters 236(2), 112–116.

iordano, J., 2005. The neurobiology of nociceptive and anti-nociceptive systems.Pain Physician 8 (3), 277–290.

upta, L.K., Gupta, R., Tripathi, C.D., 2011. N-Methyl-d-aspartate receptor modu-lators block hyperalgesia induced by acute low-dose morphine. Clinical andExperimental Pharmacology and Physiology 38 (9), 592–597.

han, M.I., Walsh, D., Brito-Dellan, N., 2011. Opioid and adjuvant analgesics: com-pared and contrasted. American Journal of Hospice & Palliative Care 28 (5),378–383.

asture, S., Vinci, S., Ibba, F., Puddu, A., Marongiu, M., Murali, B., Pisanu, A.,Lecca, D., Zernig, G., Acquas, E., 2009. Withania somnifera prevents morphinewithdrawal-induced decrease in spine density in nucleus accumbens shell ofrats: a confocal laser scanning microscopy study. Neurotoxicity Research 16 (4),343–355.

issin, I., Brown, P.T., Robinson, C.A., Bradley Jr., E.L., 1991. Acute tolerance in mor-phine analgesia: continuous infusion and single injection in rats. Anesthesiology74 (1), 166–171.

issin, I., Lee, S.S., Arthur, G.R., Bradley Jr., E.L., 1997. Effect of midazolam on develop-ment of acute tolerance to alfentanil: the role of pharmacokinetic interactions.Anesthesia & Analgesia 85 (1), 182–187.

issin, I., Bright, C.A., Bradley Jr., E.L., 2000. The effect of ketamine on opioid-induced

acute tolerance: can it explain reduction of opioid consumption with ketamine-opioid analgesic combinations? Anesthesia & Analgesia 91 (6), 1483–1488.

ozela, E., Danysz, W., Popik, P., 2001. Uncompetitive NMDA receptor antagonistspotentiate morphine antinociception recorded from the tail but not from thehind paw in rats. European Journal of Pharmacology 423 (1), 17–26.

e 21 (2014) 745–752

Kulkarni, S.K., Ninan, I., 1997. Inhibition of morphine tolerance and depend-ence by Withania somnifera in mice. Journal of Ethnopharmacology 57 (3),213–217.

Low, Y., Clarke, C.F., Huh, B.K., 2012. Opioid-induced hyperalgesia: a review of epi-demiology, mechanisms and management. Singapore Medical Journal 53 (5),357–360.

Low Dog, T., 2008. Botanicals in the management of pain. In: Audette, J.F., Bailey, A.(Eds.), Contemporary Pain Medicine: The Science and Practice of Contemporaryand Alternative Medicine in Pain Management. Umana Press, Totowa, NJ, pp.447–470.

Mehta, A.K., Binkley, P., Gandhi, S.S., Ticku, M.K., 1991. Pharmacological effects ofWithania somnifera root extract on GABAA receptor complex. Indian Journal ofMedical Research 94, 312–315.

Milne, B., Jhamandas, K., Sutak, M., Grenier, P., Cahill, C.M., 2013. Stereo-selectiveinhibition of spinal morphine tolerance and hyperalgesia by an ultra-low doseof the alpha-2-adrenoceptor antagonist efaroxan. European Journal of Pharma-cology 702 (1–3), 227–234.

Pasero, C., McCaffery, M., 2012. Opioid-induced hyperalgesia. Journal of PeriAnes-thesia Nursing 27 (1), 46–50.

Patwardhan, B., 2005. Ethnopharmacology and drug discovery. Journal ofEthnopharmacology 100 (1–2), 50–52.

Raffa, R.B., Pergolizzi Jr., J.V., 2011. Opioid-induced hyperalgesia: is it clinically rel-evant for the treatment of pain patients? Pain Management Nursing, 1–17.

Ruiu, S., Pinna, G.A., Marchese, G., Mussinu, J.M., Saba, P., Tambaro, S., Casti, P., Vargiu,R., Pani, L., 2003. Synthesis and characterization of NESS 0327: a novel puta-tive antagonist of the CB1 cannabinoid receptor. Journal of Pharmacology andExperimental Therapeutics 306 (1), 363–370.

Ruiu, S., Longoni, R., Spina, L., Orrù, A., Cottiglia, F., Collu, M., Kasture, S., Acquas,E., 2013. Withania somnifera prevents acquisition and expression of morphine-elicited conditioned place preference. Behavioral Pharmacology 24 (2),133–143.

Shimizu, N., Kishioka, S., Maeda, T., Fukazawa, Y., Yamamoto, C., Ozaki, M.,Yamamoto, H., 2004. Role of pharmacokinetic effects in the potentiation ofmorphine analgesia by L-type calcium channel blockers in mice. Journal of Phar-macological Sciences 94 (3), 240–245.

Tejwani, G.A., Rattan, A.K., Sribanditmongkol, P., Sheu, M.J., Zuniga, J., McDonald, J.S.,1993. Inhibition of morphine-induced tolerance and dependence by a benzodi-azepine receptor agonist midazolam in the rat. Anesthesia & Analgesia 76 (5),1052–1060.

Vinik, H.R., Kissin, I., 1998. Rapid development of tolerance to analgesia duringremifentanil infusion in humans. Anesthesia & Analgesia 86 (6), 1307–1311.

Wong, E.H., Kemp, J.A., Priestley, T., Knight, A.R., Woodruff, G.N., Iversen, L.L.,1986. The anticonvulsant MK-801 is a potent N-methyl-d-aspartate antagonist.

Proceedings of the National Academy of Sciences 83 (18), 7104–7108.

Zhao, M., Joo, D.T., 2008. Enhancement of spinal N-methyl-d-aspartate recep-tor function by remifentanil action at delta-opioid receptors as a mechanismfor acute opioid-induced hyperalgesia or tolerance. Anesthesiology 109 (2),308–317.