gy 567 (2007) 125–130www.elsevier.com/locate/ejphar

European Journal of Pharmacolo

Synergy between Δ9-tetrahydrocannabinol and morphine in the arthritic rat

Melinda L. Cox, Victoria L. Haller ⁎, Sandra P. Welch

Department of Pharmacology and Toxicology, Virginia Commonwealth University, Richmond, Virginia 23298-0524, United States

Received 8 February 2007; received in revised form 27 March 2007; accepted 1 April 2007Available online 20 April 2007

Abstract

We have shown in past isobolographic studies that a small amount of Δ9-tetrahydrocannabinol (Δ9-THC) can enhance morphineantinociception in mice. However, previous studies of the Δ9-THC/morphine interaction were performed using normal mice or rats and evaluatedacute thermal antinociception. Less is known about cannabinoid and opioid interactions involved in mechanical nociception and in chronicinflammatory pain models, such as Freund's complete adjuvant-induced arthritic model. One fixed-ratio combination was chosen for testing theinteraction between Δ9-THC and morphine in the Freund's adjuvant-induced arthritic model. This combination represented a 1:1 ratio of the drugsand thus consisted of equieffective doses ranging from 0.1 to 5 mg/kg Δ9-THC and from 0.1 to 5 mg/kg morphine. The combination ED50 valuefor the fixed ratios (total dose) in relation to the ED50 value of the drugs alone was determined. The isobolographic analysis indicated a synergisticinteraction between Δ9-THC and morphine in both the non-arthritic and the arthritic rats. Since Freund's adjuvant-induced alteration inendogenous opioid tone has been previously shown, our data indicate that such changes did not preclude the use of Δ9-THC and morphine incombination. As with acute preclinical pain models in which the Δ9-THC/morphine combination results in less tolerance development, theimplication of the study for chronic pain conditions is discussed.Published by Elsevier B.V.

Keywords: Isobologram; Synergy; Arthritis; Δ9-THC; Morphine

1. Introduction

Cannabinoids and opioids share several pharmacologicaleffects, including hypothermia, sedation, analgesia, and theinhibition of motor activity (Bloom and Dewey 1978; Manza-nares et al., 1999) through actions at cannabinoid and opioidreceptors, members of the G-protein-coupled receptor family.Activation of G-protein-coupled receptors produces intracellularevents such as inhibition of adenylate cyclase activity (Sharmaet al., 1975; Howlett and Fleming, 1984), decreased calciuminflux, and increased potassium efflux (Morita and North, 1982;Hescheler et al., 1987; Felder et al., 1992). Cannabinoid CB1

receptors and mu-opioid receptors have been reported to co-localize in brain areas involved in nociceptive responses such asthe periaqueductal grey, amygdala, and thalamus (Mansour et al.,1988; Martin et al., 1999). These structures are part of adescending pain control circuit that mediates pain suppressive

⁎ Corresponding author. P.O. Box 980524, Richmond, VA 23298, UnitedStates. Tel.: +1 804 828 8446; fax: +1 804 828 2117.

E-mail address: vhaller@vcu.edu (V.L. Haller).

0014-2999/$ - see front matter. Published by Elsevier B.V.doi:10.1016/j.ejphar.2007.04.010

actions of both opioids and cannabinoids (Basbaum and Fields,1984; Meng et al., 1998). In the spinal cord, opioid andcannabinoid receptors are co-localized in areas of the dorsalhorn where they are involved in nociceptive control (Salio et al.,2001).

Endogenous opioids have been shown to mediate cannabi-noid-induced antinociception. Δ9-THC-induced antinociceptionwas found to be modulated by mu-opioid receptors supraspinally,while kappa-opioid receptors were involved in spinal antinoci-ception (Smith et al., 1994; Reche et al., 1996). Spinallyadministered Δ9-THC releases dynorphin A in the spinal cordof the rat (Mason et al., 1999). Δ9-THC and morphineadministration by any combination of routes significantlyenhances the potency of morphine in mice (Welch and Stevens,1992; Smith et al., 1998a). A later study (Cichewicz et al., 1999)confirmed that a non-antinociceptive oral dose of Δ9-THCenhances the potency of an acute oral dose ofmorphine, as well asother opioid analgesics. Furthermore, a full isobolographicanalysis of the interaction between oral Δ9-THC and morphineor codeine provided evidence of synergy between Δ9-THC andthese opioids (Cichewicz and McCarthy, 2003).

126 M.L. Cox et al. / European Journal of Pharmacology 567 (2007) 125–130

Previous studies of the Δ9-THC/morphine interaction wereperformed using normal mice or rats and evaluated acutethermal antinociception. Less is known about cannabinoid andopioid interactions involved in mechanical nociception and inchronic inflammatory pain models, such as Freund's completeadjuvant-induced arthritic model. Freund's adjuvant treatmentproduces chronic inflammation, edema, and hyperalgesia in rats(Millan et al., 1986a). Sofia et al. (1973) demonstrated that Δ9-THC is effective in the paw-pressure test for mechanicalnociception in rats. In Freund's adjuvant-induced arthritic rats,Δ9-THC-elicited antinociceptive efficacy was no different fromthat in normal rats (Smith et al., 1998b). However, Δ9-THCmodulation of dynorphin Awas found to differ in normal versusarthritic rats. While Δ9-THC was shown to trigger an increasein release of dynorphin A in normal rats, arthritic rats had highlevels of dynorphin A and Δ9-THC normalized dynorphin Alevels to that of normal animals (Cox and Welch, 2004). Sinceendogenous opioid release has been shown to play a major rolein the enhancement of morphine by Δ9-THC in acute painmodels (Pugh et al., 1996), our study was designed to determineif this enhancement was observed in arthritic and normal ratsusing mechanical stimuli.

2. Materials and methods

2.1. Animals

Male Sprague–Dawley rats (Harlan Laboratories, Indiana-polis, IN), which weighed 350 to 375 g were housed in ananimal care facility maintained at 22±2 °C on a 12-h light/darkcycle with free access to food and water. All experiments wereconducted according to guidelines established by the Institu-tional Animal Care and Use Committee of Virginia Common-wealth University and adhere to the guidelines of theCommittee for Research and Ethical Issues of IASP.

2.2. Freund's adjuvant-induced arthritis treatment

A volume of 0.1 ml of vehicle (mineral oil) or Freund'scomplete adjuvant (heat-killed Mycobacterium butyricum;5 mg/ml) was injected intradermally into the base of the tail.Animals remained in their cages for 12 days and wereacclimated daily to paw-pressure testing until day 19, onwhich they were tested. Acclimation consisted of a briefpressure stimulus, less than that necessary to elicit pawwithdrawal, to the hind paw of the rat using the paw-pressureapparatus. Inflammation proceeds into a generalized polyar-thritis within 19 days. Paw-pressure baseline measurements onday 19 indicated that arthritic rats are more sensitive tomechanical nociception than non-arthritic rats.

2.3. Paw-pressure test

The paw-pressure test consisted of gently holding the bodyof the rat while the hind paw was exposed to increasingmechanical pressure. The Analgesy-Meter (Ugo-Basile, Varese,Italy) is designed to exert a force on the paw that increases at a

constant rate, similar to the Randall and Sellito (1957) test ofmechanical nociception. Force was applied to the hind paw thatwas placed under a small plinth under a cone-shaped plungerwith a rounded tip. The operator depressed a pedal-switch tostart the mechanism that exerted force. The force in grams atwhich the rat withdrew its paw was defined as the paw-pressurethreshold. The baseline paw-pressure was measured beforeinjecting vehicle or drug. Non-arthritic rats that had a baselinepaw-pressure greater than 100 g (average=177±6.42 g) wereused in further testing. Arthritic rats that had a baseline paw-pressure less than 100 g (average=71±3.05 g) were used infurther experimentation. The upper limit of 500 g was imposedfor the experiments to allow the foot to not become immobilizeddue to undue pressure.

2.4. Drug administration protocol

For the generation of dose–response curves using the paw-pressure test of antinociception in arthritic and non-arthritic rats,morphine and Δ9-THC were administered i.p. The dose–response curves for morphine and Δ9-THC alone weredetermined. Morphine was prepared in distilled water and Δ9-THC was prepared in a solution of emulphor, ethanol, and salineat a 1:1:18 ratio. Morphine (0.5, 1, 2, and 4 mg/kg) or distilledwater vehicle was administered 30 min prior to antinociceptivetesting. Δ9-THC (0.5, 1, 2 and 4 mg/kg) or 1:1:18 vehicle wasadministered 30 min prior to antinociceptive testing. The peaktimes for antinociception produced by Δ9-THC and morphinehave been previously determined to be 30 min post adminis-tration (Cox and Welch, 2004).

2.5. Percent maximum possible effect determination

The average paw-pressure threshold in grams was deter-mined before drug administration (baseline) and the selectedtimes (test) following drug administration. The maximumpossible effect (%MPE) was determined for each rat accordingto the following formula using a 500 g maximum pressure: %MPE= [test (g)−baseline (g) / 500 g−baseline (g)] × 100.Dose–response curves were generated using three or fourdoses of test drug. ED50 values and 95% confidence limits weredetermined using the methods of Tallarida and Murray (1987).Injection of distilled water (i.p., 0.1 cc/100 g body weight)resulted in a %MPE less than 5±1%. Injection of 1:1:18emuphor:ethanol:saline vehicle resulted in a %MPE of 8±2%.

2.6. Isobolographic analysis

The use of the isobologram has been reviewed extensively inthe context of drug combination studies (Wessinger, 1986;Tallarida et al, 1989; Tallarida, 2001). The method used in thepresent studies is similar to that reported by Kimmel et al.(1997). Dose–response curves were generated for each drugalone, and ED50 values (dose which yields 50% effect) andstandard error (S.E.M.) were computed using unweighted least-squares linear regression as modified from procedures 5 and8 described by Tallarida and Murray (1987). The ED50 values of

Fig. 1. Dose–response curve ofΔ9-THC (i.p.) in non-arthritic (indicated by clearsquares) and arthritic rats (indicated by black diamonds). Animals were treatedwith Δ9-THC (i.p.) alone were tested 30 min later using the paw-pressure test.The data are presented as %MPE±S.E.M using an N=6–8 rats per dose.Separate rats were used for each dose.

Table 1ED50 values±S.E.M. for Δ9-THC and morphine (both i.p.) in the pawwithdrawal test antinociception in non-arthritic rats and in arthritic rats

Δ9-THC (micrograms/kg) Morphine (micrograms/kg)

Non-arthritic rats Non-arthritic rats

Z1=2100±300 Z2=2400±400Log z1=3.222±(2.48) Log z2=3.38±(2.60)

Δ9-THC (micrograms/kg) Morphine (micrograms/kg)

Arthritic rats Arthritic rats

Z1=2500±500 Z2=2200±400Log z1=3.40±(2.70) Log z2=3.34±(2.60)

Amounts are total (in micrograms/kg for ease of log calculations).

127M.L. Cox et al. / European Journal of Pharmacology 567 (2007) 125–130

the drugs alone are then plotted and a theoretical additive line isconstructed on an isobologram. Experimental values from thefixed-ratio design studies were also analyzed using linearregression and an ED50 value for each combination wasdetermined and plotted on the isobologram for comparison tothe theoretical additive value. This theoretical value, termedZadd, is calculated using the formula Zadd= fz1+gz2 (Tallaridaet al., 1997, Eq. (3)), where f+g=1 (the proportions of eachdrug) and z1 and z2 represent the ED50 values for each drugalone. The standard error for Zadd is determined from theformula: S.E.M.(Zadd)= [f2{S.E.M.(z1)}2+g2{S.E.M.(z2)}2]1/2(Tallarida et al., 1997, Eq. (4)). The Student's t-test was used todetermine statistical significance of the difference between thelogarithmic equivalents of the ED50 values (since a requirementof the t-test is the use of values that are normally distributed). Amore detailed explanation of the calculations used for the t-testcan be found in the literature (Tallarida, 2001). A P value lessthan 0.05 indicated that the drugs produced a synergistic effect.

Fig. 2. Dose–response curve of morphine (i.p.) in non-arthritic (indicated byclear squares) and arthritic rats (indicated by black diamonds). Animals weretreated with morphine (i.p.) alone were tested 30 min later using the paw-pressure test. The data are presented as %MPE±S.E.M using an N=6–8 rats perdose. Separate rats were used for each dose.

2.7. Statistical analysis

The average paw-pressure threshold in grams was deter-mined before drug administration (baseline) and the selectedtimes (test) following drug administration. The maximumpossible effect (%MPE) was determined for each rat accordingto the following formula using a 500 g maximum pressure: %MPE= [test (g)−baseline (g) / 500 g−baseline (g)] × 100.Dose–response curves were generated using at least threedoses of test drug. ED50 value and 95% confidence limitswere determined using the methods of Tallarida and Murray(1987).

2.8. Drugs

Freund's complete adjuvant was prepared in mineral oilsupplied by Sigma Chemical Co. (St. Louis, MO). Freund'scomplete adjuvant contains heat-killed M. butyricum and issupplied by Difco Laboratories (Detroit, MI). Morphine andΔ9-THC were obtained from the National Institute on DrugAbuse (Rockville, MD).

Fig. 3. Isobologram of Δ9-THC-THC/morphine drug combination in non-arthritic rats. The points designated z1and z2 represent the ED50 values for eachdrug alone, and the line connecting these points contains all dose pairs that aresimply additive. Point A represents the theoretically additive value for thecombinations z1:z2. Point B represents the experimentally determined value forthe combination z1:z2. Since point B falls to left and below the line of theoreticaladditivity, synergy between Δ9-THC and morphine has occurred.

Table 2The amounts of Δ9-THC and morphine in a fixed-ratio mixture andcorresponding additive amounts

Proportion Δ9-THC/morphine

Experimental mixture Additive (theoretical)

Non-arthritics Zmix log (Zmix) Zadd log (Zadd)

1:1 400±50 2.6±0.041⁎ 2250±50 3.35±0.0525z1:z2

Proportion Δ9-THC/morphine

Experimental mixture Additive (theoretical)

Arthritics Zmix log (Zmix) Zadd log (Zadd)

1:1 600±55 2.77±0.039⁎ 2350±75 3.37±0.0527z1:z2

Amounts are total (in micrograms/kg for ease of log calculations) in the mixtureto produce 50% effect, and are given along with logarithmic values and standarderrors. ⁎Significant, Pb0.05 from corresponding additive values.

128 M.L. Cox et al. / European Journal of Pharmacology 567 (2007) 125–130

3. Results

3.1. Dose–response analysis of drugs alone

Figs. 1 and 2 show the dose–response curves for the antino-ciceptive effects of morphine and Δ9-THC respectively alone inrats. ED50 values (z1, z2) and S.E.M. for each drug, as well aslogarithmic equivalent doses, are presented in Table 1. Each of theED50 values is in accordance with earlier studies (Cox andWelch,2004). These values represent the equieffective doses of the drugsin these studies. The ED50 value for morphine with 95%confidence limits was 2.4 mg/kg (2.2–2.8) in normal rats. Inarthritic rats, the ED50 value for morphine with 95% confidencelimits was 2.2mg/kg (1.9–2.4). The ED50 value forΔ

9-THCwith95% confidence limits was 2.1 mg/kg (1.8–2.5) in normal rats. Inarthritic rats, the ED50 value for Δ

9-THC with 95% confidencelimits was 2.5 mg/kg (2.2–3.0). Thus, both drugs were equipotentand equiefficacious in the non-arthritic and the arthritic rats.

3.2. Isobolographic analysis of Δ9-THC/morphine interactions

One fixed-ratio combination was chosen for testing theinteraction between Δ9-THC and morphine. This combinationrepresented a 1:1 ratio of z1:z2 and thus consisted of equieffectivedoses ranging from 0.1 to 5mg/kgΔ9-THC and from 0.1 to 5mg/kgmorphine.We have shown in past isobolographic studies that asmall amount ofΔ9-THC can enhance morphine antinociceptionin mice (Cichewicz and McCarthy, 2003). Figs. 3 and 4 show theplots of the combination ED50 values for the fixed ratios (totaldose) in relation to the ED50 values of the drugs alone. Thetheoretical additive point for each drug combination is indicatedon the graph by A, while the experimental points for each drugcombination are indicated on the graph by B. The isobologramindicates that a synergistic interaction occurs between Δ9-THCand morphine in both the non-arthritic and the arthritic rats sincethe experimental points lie significantly below the line of

Fig. 4. Isobologram of Δ9-THC-THC/morphine drug combination in arthriticrats. The points designated z1and z2 represent the ED50 values for each drugalone, and the line connecting these points contains all dose pairs that are simplyadditive. Point A represents the theoretically additive value for the combinationsz1:z2. Point B represents the experimentally determined value for thecombination z1:z2. Since point B falls to left and below the line of theoreticaladditivity, synergy between Δ9-THC and morphine has occurred.

additivity. This graphical display of synergism is confirmedmathematically by comparison of the experimental values to thetheoretical values using the Student's t-test. Table 2 lists theexperimental and additive ED50 values and S.E.M. as well as theirlogarithmic equivalents. For the ratio tested, the experimentalvalue is less than the calculated additive value, and the differenceis statistically significant (Pb0.05); thus, the combination ofΔ9-THC and morphine shows synergism in both non-arthritic andarthritic rats.

4. Discussion

The first goal of the study was to determine if Δ9-THCenhanced morphine-induced antinociception in the paw-pressuretest in both normal and Freund's complete adjuvant-inducedarthritic rats. If an enhancement of antinociception was observed,we wanted to determine if the two drugs had a synergisticinteraction. To determine if Δ9-THC would enhance morphine-induced antinociception, a combination of a fixed low dose ofΔ9-THC (0.5 mg/kg) with low doses of morphine was tested in thepaw-pressure test. Our results indicated that in both non-arthriticand arthritic rats, the morphine-dose–response curve was shiftedto the left (data not shown). This preliminary study led us toexamine whether Δ9-THC would have an additive interactionwith morphine or a synergistic relationship using isobolographicanalyses. An additive interaction describes the simple addition ofthe expected effects of each dose of drug alone, while synergismrepresents the combined effect of the drugs, which greatlyexceeds that expected by simple addition.We found thatΔ9-THCand morphine, in a simple fixed dose ratio of 1:1 (z1:z2) displayeda synergistic interaction.

The synergistic interaction that we demonstrated waspredicted by previous findings. Early studies by Ghosh andBhattacharya (1979) in rats demonstrated the enhancement ofmorphine injected i.p. by an extract of Cannabis indica by thesame route. Several years later, the potency of codeine andmorphine given orally was found to be enhanced by orallyadministered Δ9-THC and Δ6-THC in mice (Mechoulam et al.,1984). More recently, low doses of Δ9-THC were found to

129M.L. Cox et al. / European Journal of Pharmacology 567 (2007) 125–130

significantly enhance morphine in the tail-flick test in the mousewhen both were administered intrathecally (Welch and Stevens,1992), and i.c.v. administration of both drugs resulted in anenhancement of morphine-induced antinociception (Welchet al., 1995). It was further established that Δ9-THC enhancedthe potency of morphine in any combination of s.c. and p.o. inmice in the tail-flick test, and that both given s.c. enhanced thepotency of morphine in mice in paw withdrawal to radiant heat(Smith et al., 1998a). The enhancement of both morphine andcodeine byΔ9-THC, administered p.o., was found to be not justadditive but synergistic interactions, determined by isobolo-graphic analysis (Cichewicz and McCarthy, 2003).

The mechanisms of the enhancement of morphine by Δ9-THC have also been examined. Naloxone (s.c.) was shown toblock the antinociceptive effects of the combination of i.t. Δ9-THC and morphine, while having no effect on Δ9-THC alone(Welch and Stevens, 1992). Antinociception produced byintrathecally administered Δ9-THC and morphine was attenu-ated by the kappa-opioid receptor antagonist, norBNI, and thedelta-opioid receptor antagonist, naltrindole. Administration ofa cocktail of enzyme inhibitors used to prevent the metabolismof dynorphin A (1–8) into leucine-enkephalin also attenuatedthe enhancement of morphine-induced antinociception by Δ9-THC (Pugh et al., 1996). Taken together these data imply a roleof delta- and kappa-opioids in the enhancement of morphineantinociception by Δ9-THC in the spinal cord. The hypothesisput forth is that the delta- and kappa-opioid interaction involvesthe Δ9-THC-induced release of dynorphin and its metabolismto leu-enkephalin, and that mu-, kappa-, and delta-opioidreceptors are required for a maximum antinociceptive effect ofthe opioid and cannabinoid combination.

These previous findings led us to hypothesize that Δ9-THCwould enhance morphine, both administered parenterally, in thepaw-pressure test in the rat. However, we also wanted toexamine this interaction in a Freund's adjuvant-induced arthriticrat. The endogenous opioid system has been shown to be alteredin the arthritic rat. Rats with chronic pain of polyarthritis exhibitalterations in the activity of the central nervous system (CNS)and hypophyseal pools of β-endorphin, met-enkephalin, anddynorphin (Millan et al., 1985, 1986a,b). Increases in tissuelevels of dynorphin and met-enkephalin and mRNA encodingtheir precursors in the dorsal horn of the spinal cord have beenreported. While tissue and mRNA levels correlate with anincrease in spinal release of dynorphin A, a decrease in met-enkrelease was observed (Millan et al., 1986a; Pohl et al., 1997;Ballet et al., 2000). We have demonstrated that whileΔ9-THC i.p. triggered a spinal release of dynorphin A in a normal rat,consistent with Mason et al. (1999), Δ9-THC administrationresulted in a lesser amount of dynorphin A in the spinal cord ofarthritic rats than that after vehicle treatment (Cox and Welch,2004). At the same time, however, Δ9-THC-induced anti-nociception was found to be no different in non-arthritic versusarthritic rats (Smith et al., 1998b; Cox and Welch, 2004). In asimilar manner, the degree of synergy of the dose combinationofΔ9-THC and morphine in this study was found to be the samein both non-arthritic and arthritic rats. Given that Δ9-THC inarthritic rats decreases dynorphin A levels, but releases

dynorphin A in non-arthritic rats (Cox and Welch, 2004), themechanism by whichΔ9-THC and morphine act synergisticallyin the arthritic rat remains unclear.

However, several important and clinically relevant pointsresult from this study: 1) Δ9-THC is a potent antinociceptiveagent in chronic Freund's complete adjuvant-induced pain. Δ9-THC is equipotent and equiefficacious to morphine in this testsystem. In other models of acute pain in humans or animals,Δ9-THC lacks the potency and efficacy of morphine (Buxbaum,1972; Bloom et al., 1977; Smith et al., 1998a); 2). TheΔ9-THC/morphine combination results in synergistic interaction in bothnormal and arthritic rats. Thus, the Freund's adjuvant-inducedalteration in endogenous opioid tone does not preclude the useof Δ9-THC and morphine in combination; 3) as with acutepreclinical pain models in which the Δ9-THC/morphinecombination results in less tolerance development, it is likelythat such might be the case in long-term treatment of chronicpain. In summary, the synergistic interaction of Δ9-THC andmorphine in arthritic rats, as well as the potency of Δ9-THC inthe CFA chronic pain condition, points to the possibility of theuse of the combination, or to the use of a Δ9-THC analog, totreat chronic pain conditions for which opioids alone areineffective.

Acknowledgement

This work was supported by National Institute on DrugAbuse Grants DA-05274, DA-07027, and 5P01DA-09789.

References

Ballet, S., Mauborgne, A., Hamon, M., Cesselin, F., Collin, E., 2000. Alteredopioid-mediated control of the spinal release of dynorphin and met-enkephalin in polyarthritic rats. Synapse 37, 262–272.

Basbaum, A.I., Fields, H.L., 1984. Endogenous pain control systems: brainstemspinal pathways and endorphin circuitry. Annu. Rev. Neurosci. 7, 309–338.

Bloom, A.S., Dewey, W.L., 1978. A comparison of some pharmacologicalactions of morphine and delta9-tetrahydrocannabinol in the mouse.Psychopharmacology (Berl) 57, 243–248.

Bloom, A.S., Dewey, W.L., Harris, L.S., Brosius, K.K., 1977. 9-nor-9beta-hydroxyhexahydrocannabinol, a cannabinoid with potent antinociceptiveactivity: comparisons with morphine. J. Pharmacol. Exp. Ther. 200, 263–270.

Buxbaum, D.M., 1972. Analgesic activity of 9-tetrahydrocannabinol in the ratand mouse. Psychopharmacologia 25, 275–280.

Cichewicz, D.L., McCarthy, E.A., 2003. Antinociceptive synergy between delta(9)-tetrahydrocannabinol and opioids after oral administration. J. Pharmacol.Exp. Ther. 304, 1010–1015.

Cichewicz, D.L., Martin, Z.L., Smith, F.L., Welch, S.P., 1999. Enhancement ofμ opioid antinociception by oral Δ9-tetrahydrocannabinol: dose–responseanalysis and receptor identification. J. Pharmacol. Exp. Ther. 289, 859–867.

Cox, M.L., Welch, S.P., 2004. The antinociceptive effect of delta9-tetrahydro-cannabinol in the arthritic rat. Eur. J. Pharmacol. 493, 65–74.

Felder, C.C., Veluz, J.S., Williams, H.L., Briley, E.M., Matsuda, L.A., 1992.Cannabinoid agonists stimulate both receptor- and non-receptor-mediatedsignal transduction pathways in cells transfected with and expressingcannabinoid receptor clones. Mol. Pharmacol. 42, 838–845.

Ghosh, P., Bhattacharya, S.K., 1979. Cannabis-induced potentiation ofmorphine analgesia in rat: role of brain monoamines. Indian J. Med. Res.70, 275–280.

Hescheler, J., Rosenthal, W., Trautwein, W., Schultz, G., 1987. The GTP-binding protein, Go, regulates neuronal calcium channels. Nature 325,445–447.

130 M.L. Cox et al. / European Journal of Pharmacology 567 (2007) 125–130

Howlett, A.C., Fleming, R.M., 1984. Cannabinoid inhibition of adenylatecyclase. Pharmacology of the response in neuroblastoma cell membranes.Mol. Pharmacol. 26, 532–538.

Kimmel, H.L., Tallarida, R.J., Holtzman, S.G., 1997. Synergism betweenbuprenorphine and cocaine on the rotational behavior of the nigrally-lesioned rat. Psychopharmacology (Berl) 133 (4), 372–377.

Mansour, A., Khachaturian, H., Lewis, M.E., Akil, H., Watson, S.J., 1988.Anatomy of CNS opioid receptors. Trends Neurosci. 11, 308–314.

Manzanares, J., Corchero, J., Romero, J., Fernandez-Ruiz, J.J., Ramos, J.A.,Fuentes, J.A., 1999. Pharmacological and biochemical interactions betweenopioids and cannabinoids. Trends Pharmacol. Sci. 20, 287–294.

Martin, W.J., Coffin, P.O., Attias, E., Balinsky, M., Tsou, K., Walker, J.M.,1999. Anatomical basis for cannabinoid-induced antinociception as revealedby intracerebral microinjections. Brain Res. 822, 237–242.

Mason Jr., D.J., Lowe, J., Welch, S.P., 1999. Cannabinoid modulation ofdynorphin A: correlation to cannabinoid-induced antinociception. Eur. J.Pharmacol 378, 237–248.

Mechoulam, R., Lander, N., Srebnik, M., Zamir, I., Breuer, A., Shalita, B.,Dikstein, S., Carlini, E.A., Leite, J.R., Edery, H., Porath, G., 1984. Recentadvances in the use of cannabinoids as therapeutic agents. In: Agurell, S.,Dewey,W.,Willette, R. (Eds.), The Cannabinoids: Chemical, Pharmacologicaland Therapeutic Aspects. Academic Press, New York, NY, pp. 777–793.

Meng, I.D., Manning, B.H., Martin, W.J., Fields, H.L., 1998. An analgesiacircuit activated by cannabinoids. Nature 395, 381–383.

Millan, M.J., Millan, M.H., Pilcher, C.W., Czlonkowski, A., Herz, A., Colpaert,F.C., 1985. Spinal cord dynorphin may modulate nociception via a kappa-opioid receptor in chronic arthritic rats. Brain Res. 340, 156–159.

Millan, M.J., Millan, M.H., Czlonkowski, A., Höllt, V., Pilcher, C.W., Herz, A.,Colpaert, F.C., 1986a. Amodel of chronic pain in the rat: response ofmultipleopioid systems to adjuvant-induced arthritis. J. Neurosci. 6, 899–906.

Millan, M.J., Millan, M.H., Czlonkowski, A., Höllt, V., Pilcher, C.W., Herz, A.,Colpaert, F.C., 1986b. Functional response of multiple opioid systems tochronic arthritic pain in the rat. Ann. N.Y. Acad. Sci. 467, 182–193.

Morita, K., North, R.A., 1982. Opiate activation of potassium conductance inmyenteric neurons: inhibition by calcium ion. Brain Res. 242, 145–150.

Pohl, M., Ballet, S., Collin, E., Mauborgne, A., Bourgoin, S., Benoliel, J.J.,Hamon, M., Cesselin, F., 1997. Enkephalinergic and dynorphinergicneurons in the spinal cord and dorsal root ganglia of the polyarthritic rat.In vivo release and cDNA hybridization studies. Brain Res. 749, 18–28.

Pugh Jr., G., Smith, P.B., Dombrowski, D.S., Welch, S.P., 1996. The role ofendogenous opioids in enhancing the antinociception produced by thecombination of delta 9-tetrahydrocannabinol and morphine in the spinalcord. J. Pharmacol. Exp. Ther. 1279, 608–616.

Randall, L.O., Sellito, J.J., 1957. A method for measurement of analgesicactivity on inflamed tissue. Arch. Int. Pharmacodyn. 111, 409–419.

Reche, I., Fuentes, J.A., Ruiz-Gayo, M., 1996. Potentiation of delta 9-tetrahydrocannabinol-induced analgesia by morphine in mice: involvementof mu-and kappa-opioid receptors. Eur. J. Pharmacol. 318, 11–16.

Salio, C., Fischer, J., Franzoni, M.F., Mackie, K., Kaneko, T., Conrath, M.,2001. CB1-cannabinoid and mu-opioid receptor co-localization on post-synaptic target in the rat dorsal horn. Neuroreport 12, 3689–3692.

Sharma, S.K., Klee, W.A., Nirenberg, M., 1975. Dual regulation of adenylatecyclase accounts for narcotic dependence and tolerance. Proc. Natl. Acad.Sci. U. S. A. 72, 3092–3096.

Sofia, R.D., Nalepa, S.D., Harakal, J.J., Vassar, H.B., 1973. Anti-edema andanalgesic properties of D9-tetrahydrocannabinol (THC). J. Pharmacol. Exp.Ther. 186, 646–655.

Smith, P.B., Welch, S.P., Martin, B.R., 1994. Interactions between delta 9-tetrahydrocannabinol and kappa opioids in mice. J. Pharmacol. Exp. Ther.268, 1381–1387.

Smith, F.L., Cichewicz, D., Martin, Z.L., Welch, S.P., 1998a. The enhancementof morphine antinociception in mice by delta9-tetrahydrocannabinol.Pharmacol. Biochem. Behav. 60, 559–566.

Smith, F.L., Fujimori, K., Lowe, J., Welch, S.P., 1998b. Characterization of D9-tetrahydrocannabinol and anandamide antinociception in nonarthritic andarthritic rats. Pharmacol. Biochem. Behav. 60, 183–191.

Tallarida, R.J., 2001. Drug synergism: its detection and applications. J. Pharmacol.Exp. Ther. 298 (3), 865–872.

Tallarida, R.J., Murray, R.B., 1987. Manual of Pharmacologic Calculations withComputer Programs. Springer-Verlag, New York, NY.

Tallarida, R.J., Porreca, F., Cowen, A., 1989. Statistical analysis of drug–drugand site–site interactions with isobolograms. Life Sci. 45, 947.

Tallarida, R.J., Stone Jr., D.J., Raffa, R.B., 1997. Efficient designs for studyingsynergistic drug combinations. Life Sci. 61 (26), 417–425.

Welch, S.P., Stevens, D.L., 1992. Antinociceptive activity of intrathecallyadministered cannabinoids alone, and in combination with morphine, inmice. J. Pharmacol. Exp. Ther. 262, 10–18.

Welch, S.P., Thomas, C., Patrick, G.S., 1995. Modulation of cannabinoid-induced antinociception following intracerebroventricular versus intrathecaladministration to mice: possible mechanisms for interaction with morphine.J. Pharmacol. Exp. Ther. 272, 310–321.

Wessinger, W.D., 1986. Approaches to the study of drug interactions inbehavorial pharmacology. Neurosci. Biobehav. Rev. 10 (2), 103–113.