Endogenous Opioid and Cannabinoid Mechanisms ...

Original Paper Received: November 1, 2011 Accepted after revision: January 1, 2012 Published online: March 10, 2012

Pharmacology 2012;89:127–136 DOI: 10.1159/000336346

© Free Author Endogenous Opioid and Cannabinoid Copy – for perMechanisms Are Involved in the Analgesic Effects sonal use only DISTRIBUTION OF THIS of Celecoxib in the Central NervousANY System ARTICLE WITHOUT WRITTEN R.M. Rezende a P. Paiva-Lima a W.G.P. Dos Reis a V.M. Y.S. Bakhle c J.N. Francischi a

CONSENT FROM S. KARGER AG, BASEL ISaA VIOLATION Camêlo A. Faraco b OF THE COPYRIGHT.

Written permission to distribute the PDF will be granted a Laboratory of Inflammation and Pain, Department of Pharmacology, Institute of Biological against payment Sciences, of a per- and b Laboratory of Pharmaceutical Products, Pharmacy School, Federal University of Minas Gerais, mission fee, which is based Belo Horizonte, Brazil; c Leukocyte Biology, National Heart and Lung Institute, Faculty of Medicine, on the number of accessesImperial College, London, UK required. Please contact [email protected]

Key Words Celecoxib ⴢ Hypoalgesia ⴢ Opioids ⴢ Cannabinoid receptor ⴢ Fatty acid amide hydrolase ⴢ Central nervous system

Abstract Background/Aims: In this study we analyzed the mechanisms underlying celecoxib-induced analgesia in a model of inflammatory pain in rats, using the intracerebroventricular (i.c.v.) administration of selective opioid and cannabinoid antagonists. Methods and Results: Analgesic effects of celecoxib were prevented by selective ␮-(␤-funaltrexamine) and ␦ -(naltrindole), but not ␬-(nor-binaltorphimine) opioid antagonists, given i.c.v. 30 min before celecoxib. Similar pretreatment with AM 251, but not SR 144528, cannabinoid CB1 and CB2 receptor antagonists, respectively, prevented celecoxib-induced analgesia. The fatty acid amide hydrolase inhibitor, URB 597, also prevented celecoxib-induced analgesia. Conclusions: Our data provided further evidence for the involvement of endogenous opioids and revealed a new cannabinoid component of the mechanism(s) underlying celecoxib-induced analgesia. Copyright © 2012 S. Karger AG, Basel

© 2012 S. Karger AG, Basel 0031–7012/12/0894–0127$38.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/pha

Introduction © Free Author

Copy – for personal use only

ANY DISTRIBUTION OF THIS ARTICLE WITHOUT WRITTEN CONSENT FROM S. KARGER AG, BASEL

TheWritten selective cyclooxygenase (COX)-2 inhibitor permission to distribute the PDF will be granted against paymentceleof a permission fee, wh coxib has been extensively used in the treatment of osteoarthritis and rheumatoid arthritis, with less gastrointestinal toxicity than the ‘classic’ nonsteroidal anti-inflammatory drugs (NSAIDs) [1, 2]. This compound and other selective inhibitors of COX-2 (hereafter referred to as ‘coxibs’) exhibit the 3 characteristic biological activities observed for classic NSAIDs – antipyretic, anti-inflammatory and analgesic [3] – attributed to their inhibition of prostaglandin biosynthesis [4]. However, in a well-established model of inflammatory pain induced by carrageenan in rat paws [5, 6], we found that celecoxib given systemically (s.c. injection) exerted antinociceptive actions that differed from those of the classic NSAIDs, such as indomethacin or piroxicam [7]. In particular, the analgesic actions of celecoxib were, unlike those of indomethacin, sensitive to blockade by opioid receptor antagonists [8]. This, along with other differences, has led us to propose that the analgesic actions of celecoxib we observed did not involve the inhibition of COX-2. Comparable analgesic effects in this model of inflammatory pain Prof. Janetti N. de Francischi Departamento de Farmacologia, ICB-UFMG Av. Antônio Carlos 6627 Pampulha, Belo Horizonte, MG 31901-270 (Brasil) Tel. +55 31 3409 2715, E-Mail janettif @ icb.ufmg.br

were also observed after giving celecoxib or its congener SC 236 to the central nervous system (CNS), by injection into the cerebral ventricles [9]. We analyzed further this central analgesic effect of celecoxib in terms of possible mechanisms, including the inhibition of COX-2. We used a structural analog of celecoxib which does not inhibit COX-2 and assessed the contribution of two other endogenous systems, opioid and endocannabinoid, to the analgesic effects of celecoxib. Our results confirmed that COX-2 inhibition is not necessary for the characteristic analgesic effects, strengthened the evidence for the participation of endogenous opioids and disclosed a new and crucial role played by cannabinoid (CB) CB1 receptors in celecoxib-induced analgesia.

Material and Methods Animals All animal care and experimental procedures adhered to the guidelines of the Committee for Research and Ethical Issues of IASP [10] and were approved by the local Ethics Committee for Animal Experimentation. Male Holtzman rats (total = 194), weighing 170–200 g and supplied by the Bioterism Center of the Federal University of Minas Gerais, were used in these experiments. After implantation of the cannula in the right lateral cerebral ventricle (see below), animals were housed one per cage with food and water ad libitum, with light/dark cycles of 12/12 h, starting at 7.00 a.m. The animals used were killed humanely at the end of the 6-hour experimental period. Reagents We used celecoxib (Celebra쏐; Pfizer Pharmaceuticals LLC – Caguas, Puerto Rico), ␤-funaltrexamine, naltrindole, nor-binaltorphimine and AM 251 (Tocris, UK), SR 144528 (a generous gift from Sanofi-Aventis, France), bestatin (Genaxxon Bioscience, Germany), URB 597 (Cayman Chemical, USA) and ␭-carrageenan (Sigma, USA). Cannulation of the Right Lateral Cerebral Ventricle Intracerebroventricular (i.c.v.) injection was performed as before [9]. Briefly, under anesthesia induced by ketamine and xylazine (10 and 2%, respectively; 1 ml/kg, i.p.-supplemented when necessary) unilateral stainless steel guide cannulas (22-gauge, 16mm length) were placed within the right lateral cerebral ventricle of rats. Animals were positioned in a stereotaxic frame (Kopf Instruments, USA) with the tooth bar fixed 5.0 mm above the level of the interaural line. Guide cannulas were positioned using the bregma as reference point and according to coordinates adapted from the atlas of Paxinos and Watson [11]: posterior –1.5 mm, lateral –2.5 mm and ventral 3.0–3.4 mm below the skull surface. The guide cannulas were fixed with dental acrylic cement and anchored by screws placed in the skull. After the surgical procedures, the animals were closely monitored for 4 h and used in experiments 1 week later.

128

Pharmacology 2012;89:127–136

Intracerebroventricular Injections Microinjections of drugs were performed with a 30-gauge injection needle (17-mm length), connected to a polyethylene tubing (0.010 ID, Norton, USA) attached to a 10-␮l Hamilton syringe (Reno, USA). All i.c.v. injections were made in a volume of 5 ␮l. Celecoxib (5.5–44 ␮g [9]) was diluted in sterile saline and cautiously injected into conscious animals 30 min before intraplantar injection of carrageenan. ␤-Funaltrexamine, naltrindole, nor-binaltorphimine (5 ␮g each [12, 13]), naltrexone (4 ␮g [9]) and bestatin (40 ␮g [9]) diluted in sterile saline, and AM 251 (10 ␮g [14]), SR 144528 (10 ␮g [15]) and URB 597 (0.01, 0.1 and 1 ␮g [16]) diluted in DMSO sterile saline (30–70%) were cautiously injected i.c.v. 30 min before celecoxib. Control animals received the same volume of the corresponding drug vehicle. At the end of the experiments, rats were injected i.c.v. with 5% Evans Blue (5 ␮l) and then killed by cervical dislocation for confirmation of the appropriate location of the cannula. If a cannula was incorrectly positioned, the result from that particular animal was excluded. Induction of Paw Inflammation Inflammation was induced in one hind paw (right) using ␭carrageenan (250 ␮g in 0.1 ml) diluted in sterile physiological saline (NaCl 0.9%) as the proinflammatory stimulus at time zero. The contralateral paw was injected with the same volume of physiological saline (0.1 ml). In earlier studies, this dose of carrageenan was shown to induce paw hyperalgesia [7, 9, 17, 18]. Measurement of Nociceptive Threshold in Paws The assessment of nociception consisted of measurement of the threshold stimulus for nociceptive reaction (paw withdrawal) using a weight (with a maximum limit of 500 g) applied to the pads of hind paws by an experimenter using a Ugo Basile apparatus, based on the Randall-Selitto method [19]. The threshold for eliciting a nociceptive response was measured before (time zero) and at 0.5, 1, 2, 3, 4 and 6 h after the intraplantar injection of carrageenan. Results are presented as the absolute values of nociceptive threshold found for each hind paw at the indicated timepoints or as the area under the curves (AUC) obtained by the trapezoidal rule from the time courses over the 2-hour interval (right/ left paw values). Thus, a fall in nociceptive thresholds (hyperalgesia) would generate a negative AUC and hypoalgesia would generate positive AUC values. Systemic AM 251 Treatment In one set of experiments, the CB1 receptor antagonist, AM 251, was also administered subcutaneously (2 mg/kg, 0.1 ml/ 100 g of weight [20]) 30 min before or after i.c.v. injection of celecoxib. Statistical Analysis Results are presented as the mean values (8SEM) obtained from groups of 4–5 animals for each condition. Differences between mean values were considered statistically significant when comparisons between the control (vehicle + carrageenan) and treated (drug + carrageenan) animals – using 1-way ANOVA, following Bonferroni’s post hoc test – gave p values !0.05.

Rezende /Paiva-Lima /Dos Reis /Camêlo / Faraco /Bakhle /Francischi

*

150

*

100

C

*

CX

50 0

5.5 11

22

44

μg/paw

–50 –100 –150

Nociceptive threshold (g)

Nociceptive threshold (AUC; g × t)

300 250

Results

#

#

#

C OSU 22 CX 22

150

*

100

*

50 0 0

Fig. 1. Effect of celecoxib (given i.c.v.) on the hyperalgesia induced by carrageenan in rat paws. Results are presented as AUC of nociceptive threshold versus time, over 2 h, calculated using the trapezoidal rule (means 8 SEM). Negative values of AUC represent hyperalgesia (a fall in nociceptive threshold) and positive values represent hypoalgesia (a threshold raised above basal). All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l) in the left paw. Animals were pretreated (i.c.v. injection; 5 ␮l) with either vehicle (C) or celecoxib (CX; 5.5, 11, 22 or 44 ␮g) 30 min before carrageenan. The data show a dose-related reversal of hyperalgesia and induction of hypoalgesia by celecoxib. * Significantly different from vehicle (C) values. # Significantly raised above basal threshold: 1-way ANOVA, p ! 0.05.

# 200

1

2

3 4 Time (h)

5

6

Fig. 2. Time course of the effect of the celecoxib analog, OSU

03012, injected i.c.v. on the hyperalgesia induced by carrageenan in rat paws. All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l) in the left paw. Animals were pretreated (i.c.v. injection; 5 ␮l) with either vehicle (C) or OSU 03012 (OSU; 22 ␮g) 30 min before carrageenan. Nociceptive thresholds (mean 8 SEM) were measured at the times indicated, for 6 h after carrageenan injection. Centrally injected OSU 03012 induced hypoalgesia for at least 1 h, as did celecoxib (22 ␮g, i.c.v.), shown here for comparison. * Significantly different from vehicle (C) values. # Significantly raised above basal threshold: 1-way ANOVA, p ! 0.05.

induced unilateral hypoalgesia, i.e. in the inflamed paw only, subsequent results will show only data from the right, inflamed paw.

Characterization of Basal Conditions: CelecoxibInduced Analgesia in Carrageenan-Injected Rat Paws Carrageenan injected into the right hind paw induced a fall in the nociceptive threshold of that paw, demonstrating the hyperalgesia characteristic of this model of inflammatory pain (see fig. 1 and 6 for AUC and other figs. for the time course). Pretreatment of animals with i.c.v. celecoxib (5.5, 11, 22 and 44 ␮g) dose-dependently raised the threshold in the carrageenan-injected (inflamed) paw to values well above that found under basal conditions (time = 0) for up to 1 h after carrageenan. An antihyperalgesic effect was maintained until the third hour after carrageenan with the higher doses of celecoxib (22 and 44 ␮g). These data are summarized as AUC in figure 1. Nociceptive thresholds were affected by celecoxib only in the inflamed paw and those in the contralateral, uninflamed paw were not changed either by the injection of carrageenan or by i.c.v. celecoxib (data not shown), confirming our previous data [9]. As celecoxib

Mechanisms of Analgesia Induced by i.c.v. Administration of Celecoxib Because we proposed that the analgesic actions of celecoxib in our model are not causally related to its inhibition of COX-2, we tested a structural analog of celecoxib, OSU 03012, lacking COX-2 inhibitory activity [21]. We injected this compound i.c.v., at a dose of 22 ␮g, 30 min before carrageenan injection to the paw, i.e. exactly the same protocol as used for celecoxib, and measured nociceptive thresholds over the next 6 h. As shown in figure 2, OSU 03012 prevented the carrageenan-induced hyperalgesia and raised the nociceptive threshold to above normal levels, a response very similar to that seen after the same dose of celecoxib (22 ␮g) in a different set of animals. Importantly, treatment with OSU 03012 did not affect the nociceptive thresholds of the contralateral, noninflamed paws, as noted above for celecoxib (data not shown).

Mechanisms of Celecoxib Analgesia


129


300 250 #

200

FNT 5 + SAL

#

SAL + CX 22 FNT 5 + CX 22

*

150

**

100

**

**

1

2

*

50 0 0

a

3 4 Time (h)

5

6


300 250

NTD 5 + SAL #

200

#

SAL + CX 22 NTD 5 + CX 22

*

150

**

100

* **

50

**

**

0 1

0

2

b

3 4 Time (h)

5

6


300 250 #

200

#

BNI 5 + SAL

#

SAL + CX 22 #

150

* *

100

BNI 5 + CX 22

*

50 0 0

c

130

1

2

3 4 Time (h)

5


6

Involvement of ␮- and ␦-Opioid Receptors in Analgesia Induced by Celecoxib We showed previously [8, 9] that celecoxib-induced analgesia in this model was prevented by the nonselective opioid receptor antagonist, naltrexone. We used selective antagonists to define more clearly the opioid receptors involved. Thus, ␤-funaltrexamine, naltrindole or norbinaltorphimine (5 ␮g each) selective for ␮-, ␦- and ␬opioid receptors, respectively, were given i.c.v. 30 min before i.c.v. injection of celecoxib. As shown in figure 3a and b, ␤-funaltrexamine or naltrindole completely prevented the analgesia induced by celecoxib, without affecting the contralateral paw (data not shown). However, nor-binaltorphimine (fig. 3c) did not modify celecoxib’s effects. Note that none of the antagonists used affected the carrageenan-induced hyperalgesia. Involvement of CBs in Celecoxib-Induced Analgesia Although the results obtained thus far supported the involvement of the opioid system in celecoxib-induced analgesia, celecoxib itself is not an agonist at opioid receptors [22] and so celecoxib may be acting indirectly, by releasing endogenous opioids. In one example of endogenous opioid release, Ibrahim et al. [23] showed that activation of CB receptors induced ␤-endorphin release. We therefore tested two CB receptor antagonists, AM 251 selective for CB1 and SR 144528 selective for CB2 receptors, in our model. AM 251 (fig. 4a), but not SR 144528 (fig. 4b), given i.c.v. 30 min before celecoxib, completely prevented the analgesic effects induced by celecoxib. Moreover, nei-

Fig. 3. Effects of selective antagonists of ␮-, ␦ - or ␬-opioid receptors on the analgesic effects of celecoxib. All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l) in the left paw. Animals were pretreated (i.c.v. injection; 5 ␮l) with the antagonist (5 ␮g or saline) 30 min before celecoxib, and with celecoxib (CX; 22 ␮g) or saline (SAL) 30 min before carrageenan. Nociceptive thresholds (mean 8 SEM) were measured at the times indicated, for 6 h after carrageenan injection. a Pretreatment with the ␮-opioid antagonist, funaltrexamine (FNT) abolished the analgesic effects of celecoxib (22 ␮g). Funaltrexamine given without celecoxib (FNT + SAL) did not affect the hyperalgesia induced by carrageenan. b Pretreatment with the ␦opioid antagonist naltrindole (NTD) also totally blocked the analgesic effects of celecoxib without affecting carrageenan-induced hyperalgesia (NTD + SAL). c By contrast, the ␬-opioid antagonist, norbinaltorphimine (BNI), did not modify celecoxib-induced hypoalgesia or the carrageenan-induced hyperalgesia. * Significantly different from corresponding values without celecoxib treatments. ** Significantly different from SAL + CX values. # Significantly raised above basal threshold: 1-way ANOVA, p ! 0.05.


300 AM 251 + SAL

250

VEH + CX 22 #

200

#

AM 251 10 + CX 22

*

150

**

100

**

* **

50



300

0

#

200

VEH + CX 22 #

150

SR 10 + SAL

#

#

* *

100

* *

SR 10 + CX 22

50 0

0

a

250

1

2

3 4 Time (h)

5

6

0

1

2

b

3 4 Time (h)

5

6

Fig. 4. Time-course of the effects of selective antagonists of CB1 or CB2 receptors on the analgesia induced by celecoxib after carrageenan. All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l) in the left paw. Animals were pretreated (i.c.v. injection; 5 ␮l) with AM 251 (10 ␮g) or SR 144528 (SR; 10 ␮g) 1 h before carrageenan, and with celecoxib (CX; 22 ␮g) 30 min before carrageenan. Nociceptive thresholds (mean 8 SEM) were measured at the times indicated, for 6 h after carrageenan injection. a Pretreatment with the CB1 receptor an-

tagonist AM 251 totally blocked the analgesic effect of celecoxib, without affecting the hyperalgesia induced by carrageenan. b However, the CB2 antagonist SR 144528 (SR) did not affect either celecoxib-induced hypoalgesia or carrageenan-induced hyperalgesia, when given alone (SR + SAL). * Significantly different from values without celecoxib treatment. ** Significantly different from VEH + CX values. # Significantly different above basal threshold: 1-way ANOVA, p ! 0.05. VEH = Vehicle.

ther antagonist given alone affected carrageenan-induced hyperalgesia (figs. 4a, b) or basal nociceptive thresholds of contralateral, noninflamed paws (data not shown). If, as suggested by these effects of AM 251, celecoxib were acting through the CB1 receptor to release endogenous opioids, then the effects of bestatin, demonstrated in our previous work [9], which depend on endogenous opioid peptides, should be affected by AM 251 as well. In the next set of experiments to test this proposition, we gave AM 251 by s.c. injection (2 mg/kg, 0.1 ml/100 g of weight) at the same time as i.c.v. bestatin (pretreatment) or 30 min after i.c.v. celecoxib injection and therefore at the same time as the intraplantar carrageenan injection (after treatment, relative to celecoxib). The combination of bestatin and low-dose celecoxib induced analgesia as expected, but this analgesia was prevented by pretreatment with AM 251 (fig. 5a). Interestingly, when the CB1 receptor antagonist was given, 30 min after celecoxib, it did not affect the analgesia induced by the combination of bestatin and low-dose celecoxib (fig. 5b).

endocannabinoids) activation of the receptor. A major endocannabinoid is anandamide, which is inactivated by the fatty acid amide hydrolase (FAAH) and inhibition of FAAH should potentiate the actions of such endogenous CBs [24–26]. We used URB 597, an irreversible inhibitor of FAAH, given i.c.v. at doses of 0.01, 0.1 or 1 ␮g (0.025– 2.5 nmol), 30 min before i.c.v. injection of celecoxib and its effects are shown in figure 6. URB 597 dose-dependently prevented celecoxib-induced analgesia, with complete inhibition at 1 ␮g. However, URB 597, at the highest dose (1 ␮g), did not affect carrageenan-induced hyperalgesia (fig. 6) or the basal nociceptive thresholds of contralateral, noninflamed paws (data not shown).

Discussion

Effect of an Irreversible Inhibitor of Fatty Acid Amide Hydrolase on Hypoalgesia Induced by Celecoxib If celecoxib was acting through the CB1 receptor, this action could represent a direct or indirect (by release of

Our results show that, in a model of inflammatory pain in rats, celecoxib given i.c.v. induced analgesia mediated by the activation of ␮- and ␦-opioid receptors and by involving the CB1 but not the CB2 receptor. The simplest explanation of these results is that the analgesic actions of celecoxib in this model were exerted indirectly by the release of endogenous opioid peptides following the activation of CB1 receptors. The results also support the con-



131

300

300 250

BE 40 + AM + CX 5.5 #

#

SAL + VEH + CX 5.5

200

BE 40 + VEH + CX 5.5

*

150 100

** **

50

**



BE 40 + AM + SAL # 200

# #

150

BE 40 + CX 5.5 + AM

#

SAL + CX 5.5 + VEH #

BE 40 + CX 5.5 + VEH

*

100 50 0

0 0

a

BE 40 + SAL + AM

250

1

2

3 4 Time (h)

5

0

6

1

2

3 4 Time (h)

b

5

6

Fig. 5. Reversal by AM 251 of celecoxib-induced analgesia is time-

tion of bestatin + celecoxib. Note that bestatin and AM 251, in the absence of celecoxib, did not modify carrageenan-induced hyperalgesia. b The same combination of bestatin and celecoxib (BE + CX 5.5 + VEH) induced hypoalgesia for at least 1 h, but when AM 251 was given (s.c.) 30 min after celecoxib, i.e 30 min later than in a, no reversal of celecoxib-induced hypoalgesia was observed. * Significantly different from values without celecoxib treatment. ** Significantly different from BE + VEH + CX values. # Significantly different above basal threshold: 1-way ANOVA, p ! 0.05. VEH = Vehicle.

cept that inhibition of COX-2 by celecoxib was not relevant to this analgesic action. We proposed a lack of causality between COX-2 inhibition and celecoxib-induced analgesia from our first experiments disclosing the characteristic hypoalgesic effects of celecoxib compared with the antihyperalgesic effects of other NSAIDs, such as indomethacin and piroxicam [7]. The divergence between COX-2 inhibition and celecoxib-induced analgesia was recently strengthened by the results with lumiracoxib, a more potent and more selective COX-2 inhibitor than celecoxib [27], which, like the classical NSAIDs, did not induce hypoalgesia but only provided antihyperalgesia [28]. Moreover, given that indomethacin and piroxicam inhibit COX-1 more potently than COX-2 [29] and that the selective COX-1 inhibitor SC 560 [30] also showed only antihyperalgesia [8], it would appear that the inhibition of prostaglandin biosynthesis in general was not a causal event in celecoxib-induced hypoalgesia. Finally, in this series of experiments, we found that OSU 03012, a structural analog of celecoxib but lacking COX inhibitory activity [21], induced analgesia identical to that induced by celecoxib. The simplest explanation of all these results is that, in our

model, celecoxib induced analgesia by a mechanism other than the inhibition of COX-2. The initial indication of opioid involvement in celecoxib-induced hypoalgesia was that it was reversed by naltrexone, whereas analgesia after indomethacin was resistant [8]. Also, chronic treatment of rats with celecoxib, but not with indomethacin, induced a tolerance to its analgesic effect [8, 31]. In addition, rats made tolerant to morphine in our model were cross-tolerant to SC 236, another structural analog of celecoxib and selective inhibitor of COX-2, and rats tolerant to celecoxib were also cross-tolerant to the analgesic effects of morphine [8, 31]. On the basis of these indications of opioid involvement in celecoxib-induced analgesia, we tested the involvement of specific opioid receptors in this effect. Either ␤-funaltrexamine or naltrindole (but not nor-binaltorphimine) was effective in preventing celecoxib-induced analgesia, suggesting that the activation of ␮- and/or ␦-opioid receptors was a critical component of celecoxib-induced analgesia. However, this activation of opioid receptors is unlikely to be a direct agonist effect on these receptors as celecoxib is not a ligand for opioid receptors [22]. This would therefore imply an indirect activation, i.e. the re-

dependent. All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l) in the left paw. Animals were pretreated (i.c.v. injection) with bestatin (BE; 40 ␮g) 1 h before carrageenan, and celecoxib (CX; 5.5 ␮g) 30 min before carrageenan. Nociceptive thresholds (mean 8 SEM) were measured at the times indicated, for 6 h after carrageenan injection. This lower dose of celecoxib (SAL + VEH + CX 5.5) did not affect carrageenan-induced hyperalgesia, but, combined with bestatin (BE 40 + VEH + CX 5.5), did induce hypoalgesia. AM 251 given s.c. before celecoxib abolished the analgesic effects of the combina-

132




Nociceptive threshold (AUC; g × t)

lease of endogenous opioids which would be the agonists for the ␮- or ␦-opioid receptors. The most likely opioids to fit this receptor profile are the enkephalins and ␤-endorphin and these opioid peptides are found in the periphery and in the CNS [32, 33]. The metabolism and inactivation of either is blocked by bestatin, and would thus be compatible with the potentiating effects we have observed with this peptidase inhibitor in celecoxib-induced analgesia [9]. Some years ago, the release of ␤-endorphin in rat paws by a CB agonist (AM1241) was reported by Ibrahim et al. [23]. This finding prompted us to test the involvement of the CBs in the possible release of endogenous opioids after i.c.v. celecoxib, using two selective CB receptor antagonists. Only AM 251, the CB1 receptor antagonist, prevented celecoxib-induced analgesia, clearly implying that these receptors were involved. Causal connections between the CB and opioid systems have been frequently reported [34] and anatomical studies have shown a similar distribution of opioids and CB receptors in several structures within the CNS [35–38]. Other evidence, particularly relevant in our context, includes the ability of opioid or CB antagonists to reverse opioid or CB-induced analgesia [39] and cross-tolerance between both systems [39, 40]. CB agonists, such as ⌬9-tetrahydrocannabinol and CP 55,940, a nonselective CB agonist and AM 1241, a selective CB2 agonist, all released endogenous opioids after binding at their receptors, an effect contributing to the subsequent analgesia [23, 41–45]. If the release of endogenous opioids was subsequent to the activation of CB1 receptors, was celecoxib the CB1 agonist or was there an additional step, the release of an endocannabinoid agonist at CB1 receptors? There are at least two pieces of evidence against celecoxib being the CB1 receptor agonist. First, in all the extensive work on the interactions between COX and the endocannabinoid system [46, 47], direct CB receptor agonist activity was not shown or postulated for COX inhibitors. Second, our finding, that inhibition of FAAH affected the interaction between celecoxib and the endocannabinoid system, implies strongly that the endocannabinoids themselves were the CB1 receptor agonist(s) involved. If then, celecoxib was acting indirectly to activate CB1 receptors, how was that achieved? Most of the COX-endocannabinoid interactions have been attributed to the inhibition of COX because COX (and particularly COX2) will metabolize both anandamide and 2-AG to products lacking CB receptor activity [46, 47]. Thus, inhibition of COX would have effects comparable to the inhibition of other endocannabinoid clearance mechanisms.

75

#

*

#

URB + SAL SAL + CX 22 URB 0.01 + CX 22 URB 0.1 + CX 22 URB 1 + CX 22

*

50 25 0 –25 –50 –75 –100 –125

** * **

Fig. 6. Dose-dependent effects of the FAAH inhibitor URB 597 on celecoxib-induced analgesia after carrageenan. Results are presented as AUC (means 8 SEM) of nociceptive threshold versus time, over 2 h, calculated using the trapezoidal rule. All rats were injected with carrageenan (250 ␮g in 100 ␮l) in the right paw and saline (100 ␮l in the left paw). Animals were pretreated with URB 597 (URB; i.c.v. 0.01, 0.1 or 1.0 ␮g) 1 h before carrageenan, and celecoxib (CX; i.c.v. 22 ␮g) 30 min before carrageenan. The first results (left-hand bar; URB + Sal) show that the highest dose of URB 597 (1 ␮g, i.c.v.), in the absence of celecoxib, did not affect carrageenan-induced hyperalgesia. The hypoalgesia induced by celecoxib (Sal + CX 22) was not changed by the lowest dose of URB 597 (URB 0.01 + CX 22), but higher doses of the FAAH inhibitor progressively reversed celecoxib-induced analgesia (URB 0.1 + CX 22; URB 1 + CX 22). * Significantly different from values without celecoxib treatment. ** Significantly different from SAL + CX values. # Significantly different above basal threshold: 1-way ANOVA, p ! 0.05.

However, in our model, the celecoxib-induced hypoalgesia was unconnected to COX inhibition, as already discussed above, so this mechanism would not explain our findings. We have therefore been forced to postulate the release of a CB1 receptor agonist, which in turn releases endogenous opioid to achieve the final analgesic effect. Such a proposition creates further problems. The most prevalent endocannabinoid, anandamide, is not able to stimulate synthesis or release of any type of endogenous opioid [43, 44, 48]. However, the marked effect of FAAH inhibition by URB 597 on celecoxib-induced hypoalgesia suggested that anandamide, a known substrate for FAAH, was involved. Inhibition of FAAH, the major inactivation pathway for anandamide, leads to potentiation of the effects of anandamide in the CNS [49] or in the periphery [50], but we have found URB 597 to inhibit celecoxibinduced analgesia. This particular paradox may be resolved by noting that hyperalgesia has been reported folPharmacology 2012;89:127–136

133

lowing synthetic CB receptor agonists [51] or capsaicin [52], delivered to the periaqueductal grey and that [53] found stimulation of TRPV1 channels after inhibition of FAAH in rat periaqueductal grey slices. Anandamide is an agonist at TRPV1 channels [54] and our results may thus represent an increase in the hyperalgesic pathways (mediated by TRPV1 channels), relative to the analgesic pathways, following FAAH inhibition. We cannot discount the other major endocannabinoid, 2-AG, as the CB1 receptor agonist stimulating opioid release in our experiments. Although inhibition of FAAH usually only affects anandamide and inhibition of monoacylglycerol lipase [55] usually only affects 2-AG levels [56], there are relevant reports of URB 597 increasing 2-AG levels [51, 57]. However, unlike anandamide, 2-AG is not an agonist at TRPV1 channels [54] and so the reversal of celecoxib-induced hypoalgesia by URB 597 would not be explained by increased 2-AG levels. Nevertheless, both anandamide and 2-AG can be released together [58, 59] and that might be the case in our experiments. One clear limitation of our work is that only one nociceptive assay system was used and there are many other models of inflammatory and of noninflammatory pain [60] with different mechanisms and hence the possibility of different outcomes. Our preference for our model was based on its value in predicting clinical outcomes for the anti-inflammatory aspects of NSAIDs [61], but clearly now that the mechanisms underlying NSAID action seem not to apply to celecoxib-induced analgesia, our findings need to be extended to other pain models.

We have analyzed celecoxib’s action after central administration by i.c.v. injection in these experiments but it is arguable that celecoxib has most of its actions in the periphery, especially after oral administration. Furthermore, celecoxib crosses the blood-brain barrier less extensively than other coxibs [62] making a central site of action less likely. Therefore, our present series of experiments will need to be repeated with the peripheral administration of celecoxib to confirm the role of CB receptors in celecoxib-induced analgesia. In summary, our results have strengthened the proposition that celecoxib acts, in our model, by releasing endogenous opioid peptides and subsequent activation of ␮- and/or ␦-opioid receptors, rather than by inhibiting COX-2. The results also disclosed the activation of CB1 receptors as another causal link in celecoxib-induced central analgesia. Although the direct activation of CB1 receptors by celecoxib and mechanisms involving COX inhibition are unlikely, how celecoxib induces the activation of CB1 receptors remains to be established. Nevertheless, the possibility raised by our work of a non-COXbased, new analgesic mechanism – to explain the relief of inflammatory pain by celecoxib – is clearly worth pursuing. Acknowledgements This work was supported by the Conselho Nacional de Pesquisa (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG MG).

References 1 Penning TB, Talley JJ, Bertenshaw SR, Carter JS, Collins PW, Docter S, Graneto MJ, Lee LF, Malecha JW, Miyashiro JM, Rogers RS, Rogier DJ, Yu SS, Anderson GD, Burton EG, Cogburn JN, Gregory SA, Koboldt CM, Perkins WE, Seibert K, Veenhuizen AW, Zhang YY, Isakson PC: Synthesis and biological evaluation of the 1,5-diarylpyrazole class of cyclooxygenase-2 inhibitors: identification of 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-y-1] benzenesulfonamide (SC-58635, Celecoxib). J Med Chem 1997;40: 1347–1365. 2 Deeks JJ, Smith LA, Bradley MD: Efficacy, tolerability, and upper gastrointestinal safety of celecoxib for treatment of osteoarthritis and rheumatoid arthritis: systematic review of randomised controlled trials. BMJ 2002; 325:619.

134

3 Burian M, Geisslinger G: COX-dependent mechanisms involved in the antinociceptive action of NSAIDs at central and peripheral sites. Pharmacol Ther 2005;107:139–154. 4 Vane JR: Inhibition of prostaglandin synthesis as a mechanism of action for aspirin like drugs. Nat New Biol 1971; 231:232–235. 5 Di Rosa M: Biological properties of the carrageenan. J Pharm Pharmac 1972;24:89–102. 6 Vinegar R, Truax JF, Selph JL, Johnston PR, Venable AL, McKenzie KK: Pathway to carrageenan-induced inflammation in the hind limb of the rat. Fed Proc 1987;46:118–126. 7 Francischi JN, Chaves CT, Moura ACL, Lima AS, Rocha OA, Ferreira-Alves DL, Bakhle YS: Selective inhibitors of cyclo-oxygenase-2 (COX-2) induce hypoalgesia in a rat paw model of inflammation. Br J Pharmacol 2002;137:837–844.


8 França DS, Ferreira-Alves DL, Duarte ID, Ribeiro MC, Rezende RM, Bakhle YS, Francischi JN: Endogenous opioids mediate the hypoalgesia induced by inhibitors of cyclooxygenase-2 in rat paws treated with carrageenan. Neuropharmacology 2006; 51: 37– 43. 9 Rezende RM, Dos Reis WG, Duarte ID, Lima PP, Bakhle YS, de Francischi JN: The analgesic actions of centrally administered celecoxib are mediated by endogenous opioids. Pain 2009;142:94–100. 10 Zimmermann M: Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;6:109–110. 11 Paxinos G, Watson C: The Rat Brain in Stereotaxic Coordinates, ed 2. Sydney, Academic Press, 1986.


12 Marchand F, Ardid D, Chapuy E, Alloui A, Jourdan D, Eschalier A: Evidence for an involvement of supraspinal delta- and spinal mu-opioid receptors in the antihyperalgesic effect of chronically administered clomipramine in mononeuropathic rats. J Pharmacol Exp Ther 2003;307:268–274. 13 Gao L, Yu LC: Involvement of opioid receptors in the oxytocin-induced antinociception in the central nervous system of rats. Regul Pept 2004;120:53–58. 14 Kozan R, Ayyildiz M, Agar E: The effects of intracerebroventricular AM-251, a CB1-receptor antagonist, and ACEA, a CB1-receptor agonist, on penicillin-induced epileptiform activity in rats. Epilepsia 2009; 50: 1760–1767. 15 Benamar K, Geller EB, Adler MW: First in vivo evidence for a functional interaction between chemokine and cannabinoid systems in the brain. J Pharmacol Exp Ther 2008;325: 641–645. 16 Lisboa SF, Resstel LB, Aguiar DC, Guimarães FS: Activation of cannabinoid CB1 receptors in the dorsolateral periaqueductal gray induces anxiolytic effects in rats submitted to the Vogel conflict test. Eur J Pharmacol 2008;593:73–78. 17 Pereira LS, Ferreira-Alves DL, Resende MA, Romualdo MA, Dos Reis WGP, Lopes MT, Francischi JN: Reduced production of hyperalgesic substances by mononuclear cells from aged rats incubated with carrageenan: role of interleukin-2 and prostaglandins. Inflamm Res 2003;52:119–125. 18 Rezende RM, França DS, Menezes GB, Dos Reis WG, Bakhle YS, Francischi JN: Different mechanisms underlie the analgesic actions of paracetamol and dipyrone in a rat model of inflammatory pain. Br J Pharmacol 2008;53:760–768. 19 Randall LO, Selitto JJ: A method for measurement of analgesic activity on inflamed tissues. Arch Int Pharmacodyn Therap 1957; 111:409–419. 20 Mallet C, Daulhac L, Bonnefont J, Ledent C, Etienne M, Chapuy C, Libert F, Eschalier A: Endocannabinoid and serotoninergic systems are needed for acetaminophen-induced analgesia. Pain 2008;139:190–200. 21 Johnson AJ, Smith LL, Zhu J, Heerema NA, Jefferson S, Mone A, Grever M, Chen CS, Byrd JC: A novel celecoxib derivative, OSU03012, induces cytotoxicity in primary CLL cells and transformed B-cell lymphoma cell line via a caspase- and Bcl-2-independent mechanism. Blood 2005; 105: 2504– 2509. 22 Gregori-Puigjané E, Mestres J: A ligandbased approach to mining the chemogenomic space of drugs. Comb Chem High Throughput Screen 2008;11:669–676. 23 Ibrahim MM, Porreca F, Lai J, Albrecht PJ, Rice FL, Khodorova A, Davar G, Makriyannis A, Vanderah TW, Mata HP, Malan TP Jr: CB2 cannabinoid receptor activation produces antinociception by stimulating pe-


24

25

26

27

28

29 30

31

32

33 34 35

36

ripheral release of endogenous opioids. Proc Natl Acad Sci USA 2005;102:3093–3098. Cravatt BF, Giang DK, Mayfield SP, Boger DL, Lerner RA, Gilula NB: Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature 1996;384:83–87. Cravatt BF, Demarest K, Patricelli MP, Bracey MH, Giang DK, Martin BR, Lichtman AH: Supersensitivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase. Proc Natl Acad Sci USA 2001;98:9371–9376. Lichtman AH, Leung D, Shelton CC, Saghatelian A, Hardouin C, Boger DL, Cravatt BF: Reversible inhibitors of fatty acid amide hydrolase that promote analgesia: evidence for an unprecedented combination of potency and selectivity. J Pharmacol Exp Ther 2004; 311:441–448. Esser R, Berry C, Du Z, Dawson J, Fox A, Fujimoto RA, Haston W, Kimble EF, Koehler J, Peppard J, Quadros E, Quintavalla J, Toscano K, Urban L, Van Duzer J, Zhang X, Zhou S, Marshall PJ: Preclinical pharmacology of lumiracoxib: a novel selective inhibitor of cyclooxygenase-2. Br J Pharmacol 2005; 144: 538–550. Gassani BCA, Rezende RM, Paiva-Lima P, Ferreira-Alves DL, Bakhle YS, Francischi JN: Is the sulphonamide radical in the celecoxib molecule essential for its analgesic activity? Pharmacol Res 2010;62:439–443. Vane JR, Bakhle YS, Botting RM: Cyclooxygenase-1 and 2. Annu Rev Pharmacol Toxicol 1998;38:97–120. Smith CJ, Zhang Y, Koboldt CM, Muhammad J, Zweifel BS, Shaffer A, Talley JJ, Masferrer JL, Seibert K, Isakson PC: Pharmacological analysis of cyclooxygenase-1 in inflammation. Proc Natl Acad Sci USA 1998; 95:13313–13318. Rezende RM, Paiva-Lima P, Camêlo VM, Dos Reis WGP, Bakhle YS, Francischi JN: Celecoxib induces tolerance in a model of peripheral inflammatory pain in rats. Neuropharmacology 2010;59:551–557. Finley JCW, Lindström P, Petrusz P: Immunocytochemical localization of ␤-endorphin-containing neurons in the rat brain. Neuroendocrinology 1981;33:28–42. Smith PA, Moran TD: The nociceptin receptor as a potential target in drug design. Drugs News and Perspect 2001;14:335–345. Welch SP: Interaction of the cannabinoid and opioid systems in the modulation of nociception. Int Rev Psychiatry 2009;21:143–151. Mansour A, Khachaturian H, Lewis ME, Akil H, Watson SJ: Autoradiographic differentiation of mu, delta and kappa opioid receptors in the rat forebrain and midbrain. J Neurosci 1987;7:2445–2464. Herkenham M, Lynn AB, Johnson MR, Melvin LS, De Costa BR, Rice KC: Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study. J Neurosci 1991;11:563–583.

37 Hohmann AG, Herkenham M: Cannabinoid receptors undergo axonal flow in sensory nerves. Neuroscience 1999;92:1171–1175. 38 Salio C, Fischer J, Franzoni MF, Mackie K, Kaneko T, Conrath M: CB1-cannabinoid and ␮-opioid co-localization on postsynaptic target in the rat dorsal horn. Neuroreport 2001;12:3689–3692. 39 Cichewicz DL: Synergistic interactions between cannabinoids and opioid analgesics. Life Sci 2004;74:1317–1324. 40 Cichewicz DL, Welch SP: Modulation of oral morphine antinociceptive tolerance and naloxone-preciptated withdrawal signs by oral ⌬9-THC. J Pharmacol Exp Ther 2003; 305: 812–817. 41 Smith PB, Welch SP, Martin BR: Interactions between delta-9-tetrahydrocannabinol and kappa opioids in mice. J Pharmacol Exp Ther 1994;268:1381–1387. 42 Pugh Jr G, Mason DJ Jr, Combs V, Welchs SP: Involvement of dynorphin B in the antinociceptive effects of the cannabinoid CP55,940 in the spinal cord. J Pharmacol Exp Ther 1997;281:730–737. 43 Mason DJ Jr, Lowe J, Welch SP: Cannabinoid modulation of dynorphin A: correlation to cannabinoid-induced antinociception. Eur J Pharmacol 1999;378:237–248. 44 Houser SJ, Eads M, Embrey JP, Welch SP: Dynorphin B and spinal analgesia: induction of antinociception by the cannabinoids CP55,940, ⌬9-THC and anandamide. Brain Res 2000;857:337–342. 45 Valverde O, Noble F, Beslot F, Foumie-Zaluski MC, Roques BP: Delta 9-tetrahydrocannabinol releases and facilitates the effects of endogenous enkephalins: reduction in morphine withdrawal syndrome without changing in rewarding effect. Eur J Neurosci 2001; 13:1816–1824. 46 Fowler CJ: The contribution of cyclooxygenase-2 to endocannabinoid metabolism and action. Br J Pharmacol 2007;152:594–601. 47 Guindon J, Hohmann AG: A physiological role for endocannabinoid-derived products of cyclooxygenase-2-mediated oxidative metabolism. Br J Pharmacol 2008;153:1341–1343. 48 Fernández-Solari J, Scorticati C, Mohn C, De Laurentiis A, Billi S, Franchi A, McCann SM, Rettori V: Alcohol inhibits luteinizing hormone-releasing hormone release by activating the endocannabinoid system. Proc Natl Acad Sci USA 2004;101:3264–3268. 49 Ahn K, Johnson DS, Cravatt BF: Fatty acid amide hydrolase as a potential therapeutic target for the treatment of pain and CNS disorders. Expert Opin Drug Discov 2009; 4: 763–784. 50 Clapper JR, Moreno-Sanz G, Russo R, Guijarro A, Vacondio F, Duranti A, Tontini A, Sanchini S, Sciolino NR, Spradley JM, Hohmann AG, Calignano A, Mor M, Tarzia G, Piomelli D: Anandamide suppresses pain initiation through a peripheral endocannabinoid mechanism. Nat Neurosci 2010; 13: 1265–1270.


135

51 Maione S, Bisogno T, De Novellis V, Palazzo E, Cristino L, Valenti M, Petrosino S, Guglielmotti V, Rossi F, Di Marzo V: Elevation of endocannabinoid levels in the ventrolateral periaqueductal grey through inhibition of fatty acid amide hydrolase affects descending nociceptive pathways via both cannabinoid receptor type 1 and transient receptor potential vanilloid type-1 receptors. J Pharmacol Exp Ther 2006;316:969–982. 52 McGaraughty S, Chu KL, Bitner RS, Martino B, El Kouhen R, Han P, Nikkel AL, Burgard EC, Faltynek CR, Jarvis MF: Capsaicin infused into the PAG affects rat tail flick responses to noxious heat and alters neuronal firing in the RVM. J Neurophysiol 2003; 90: 2702–2710. 53 Kawahara H, Drew GM, Christie MJ, Vaughan CW: Inhibition of fatty acid amide hydrolase unmasks CB1 receptor and TRPV1 channel-mediated modulation of glutamatergic synaptic transmission in midbrain periaqueductal grey. Br J Pharmacol 2011; 163:1214–1222.

54 Pertwee RG, Howlett AC, Abood ME, Alexander SP, Di Marzo V, Elphick MR, Greasley PJ, Hansen HS, Kunos G, Mackie K, Mechoulam R, Ross RA: International Union of Basic and Clinical Pharmacology. LXXIX. Cannabinoid receptors and their ligands: beyond CB1 and CB2. Pharmacol Rev 2010;62:588–631. 55 Sugiura T, Kishimoto S, Oka S, Gokoh M: Biochemistry, pharmacology and physiology of 2-arachidonoylglycerol, an endogenous cannabinoid receptor ligand. Prog Lipid Res 2006;45:405–446. 56 Buczynski MW, Parsons LH: Quantification of brain endocannabinoid levels: methods, interpretations and pitfalls. Br J Pharmacol 2010;160:423–442. 57 Jhaveri MD, Richardson D, Robinson I, Garle MJ, Patel A, Sun Y, Sagar DR, Bennett AJ, Alexander SP, Kendall DA, Barrett DA, Chapman V: Inhibition of fatty acid amide hydrolase and cyclooxygenase-2 increases levels of endocannabinoid related molecules and produces analgesia via peroxisome proliferator-activated receptor-alpha in a model of inflammatory pain. Neuropharmacology 2008;55:85–93. 58 Béquet F, Uzabiaga F, Desbazeille M, Ludwiczak P, Maftouh M, Picard C, Scatton B, Lê Fur G: CB1 receptor-mediated control of the release of endocannabinoids (as assessed by microdialysis coupled with LC/MS) in the rat hypothalamus. Eur J Neurosci 2007; 26: 3458–3464.

59 Di S, Malcher-Lopes R, Marcheselli VL, Bazan NG, Tasker JG: Rapid glucocorticoidmediated endocannabinoid release and opposing regulation of glutamate and gammaaminobutyric acid inputs to hypothalamic magnocellular neurons. Endocrinology 2005;146:4292–4301. 60 Le Bars D, Gozariu M, Cadden SW: Animal models of nociception. Pharmacol Rev 2001; 53:597–652. 61 Mukherjee A, Hale VG, Borga O, Stein R: Predictability of the clinical potency of NSAIDs from the preclinical pharmacodynamics in rats. Inflamm Res 1996; 45: 531– 540. 62 Bingham S, Beswick PJ, Bountra C, Brown T, Campbell IB, Chessell IP, Clayton N, Collins SD, Davey PT, Goodland H, Gray N, Haslam C, Hatcher JP, Hunter AJ, Lucas F, Murkitt G, Naylor A, Pickup E, Sargent B, Summerfield SG, Stevens A, Stratton SC, Wiseman J: The cyclooxygenase-2 inhibitor GW406381X [2-(4-ethoxyphenyl)-3-[4-(methylsulfonyl) phenyl]-pyrazolo[1,5-b]pyridazine] is effective in animal models of neuropathic pain and central sensitization. J Pharmacol Exp Ther 2005;312:1161–1169.

© Free Author Copy – for personal use only ANY DISTRIBUTION OF THIS ARTICLE WITHOUT WRITTEN CONSENT FROM S. KARGER AG, BASEL IS A VIOLATION OF THE COPYRIGHT. Written permission to distribute the PDF will be granted against payment of a permission fee, which is based on the number of accesses required. Please contact [email protected]

136



© Free Author Copy – for personal use only ANY DISTRIBUTION OF THIS ARTICLE WITHOUT WRITTEN CONSENT FROM S. KARGER AG, BASEL IS A VIOLATION OF THE COPYRIGHT.