Contribution of metabotropic glutamate receptors to opioid dependence

Marian Elaine Fundytus

Department of Psychology

McGill, University

December, 1996

A thesis submitted to the Faculty of Graduate Studies and Research in partial fuifiilment of

the requirements of the degree Doctor of Philosophy

BMarian Elaine Fundytus, 1996 National Library Bibliothèque nationale I*I of Canada du Canada Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. rue Wellington OttawaON KlAOiU4 Ottawa ON K1A ON4 Canada Canada

The author has granted a non- L'auteur a accordé une licence non exclusive licence allowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or seU reproduire, prêter, distribuer ou copies of this thesis in microform, vendre des copies de cette thèse sous paper or electronic formats. la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts from it Ni la thèse ni des extraits substatiels may be p~tedor otheMrise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. Candidates have the option of includig, as part of the thesis, the text of a paper(s) subrnitted or to be submitted for publication, or the clearly duplicated text of a published paper(s). These texts must be bound as an integral part of the thesis.

If this option is chosen, connecting texts that provide logical bridges between the different papers are mandatory. The thesis must be written in such a way that it is more than a mere collection of manuscripts; in other words, results of a series of papers rnust be integrated.

The thesis must still conform to al1 other requirements of the "Guidelines for

Thesis Preparation". The thesis must include: A Table of Contents, an abstract in

English and French, an introduction which clearly states the rationale and objectives of the study, a comprehensive review of the literature, a hai conclusion and summary, and a thorough bibliography or reference list.

Additional matenal must be provided where appropriate (e.g. in appendices) and in sufficient detail to allow a clear and precise judgernent to be made of the importance and originality of the research reported in the thesis.

In the case of manuscripts CO-authoredby the candidate and others, the candidate is required to rnake an explicit statement in the thesis as to who contributed to such work and to what extent. SupeMsors must attest to the accuracy of such statements at the doctoral oral defense. Since the task of the examiners is made more difficult in these cases, it is in the candidate's interest to make perfectly clear the responsibilities of al1 the authors of the CO-authoredpapers. Under no circumstances can a CO-authorof any component of such a thesis serve as an examiner for that thesis. TABLE OF CONTENTS

Acknowledgements Abstract Résumé Chapter 1 : Introduction 1. Brief oveMew of opioids A. Opioid receptors and endogenous ligands B. Opioid receptor subtypes C. Tolerance and dependence i) Relationship between p and 6 receptors ii) Relationship between p and IC receptors 11. Excitatory amino acids A. Receptors i) NMDA receptors ii) AMPA receptors iii) Kainate recepton iv) Metabotropic receptors B. Excitatory amino acids and morphine tolerance and dependence III. Opioids and EUlinked second messengers A. Cyclic adenosine 3',S1-monophosphate(CAMP) production B. Phosphatidylinositol hydrolysis

Chapter 2: A closer look at the effects of excitatory amino acid antagonists on morphine tolerance and dependence

Chapter 3: A further examination of the contribution of mGluR subtypes to the development of morphine dependence

Chapter 4: Assessrnent of the relationship between products of phosphatidylinositol hydrolysis and morphine dependence

Chapter 5: Studies of the role of mGluR desensitization in opioid dependence

Chapter 6: Involvement of 6-opioid receptors in the development of morphine tolerance and dependence

Chapter 7: General Discussion 1. Surnrnary of results in relation to previous data II. Previous models of EAA receptor contributions and CAMP to morphine dependence A. NMDA receptor supersensitivity B. NMD A-induced PKC activity C. CAMP compensatory mechanisms III. A new mode1 for EAA receptor mediation of morphine dependence: criticai role of mGluRs IV. Future directions

References Appendix 1: Original contributions Appendix II: Copyright permission TABLE OF FIGURES

Figure 2.1 Figure 2.2 Figure 2.3 Figure 3.1 Figure 3.2 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 5.1 Figure 5 -2 Figure 6.1 Figure 7.1 TABLE OF TABLES

Table 6.1 Acknowledgements

I would Like to offer a special thank-you to Dr. Terence J. Coderre, my thesis supervisor, for his patience and valuable guidance d~nngthe completion of the research and writing of the dissertation. His support made the whole process a pleasure. 1 would like to thank JederRitchie for her vaiuable aid in the collection of data, both for this thesis and extracurricular projects, especiaüy when 1 was pregnant and with a young baby.

1 would like to offer a special thank-you to my husband, Douglas Jolivet. Not only was he very patient and supportive, but he made sure aii the lab equipment and cornputer prograrns were in top working condition. 1 would dso like to thank my daughter Kayla Jolivet, for being patient when her mommy couldn't spend much time with her.

I would like to thank my fiends Tammi Kwan, Lisa Weaver, Kim Fisher and

Rossanna Fargnoli for ail their support. They were always there when 1 needed someone to talk to. Abstract

We investigated the role of metabotropic glutamate receptors (mGluRs), and related intracellular second messengers, in the development of morphine tolerance and dependence. The mGluRs are divided into three groups: group 1 rnGluRs are positively coupled to phosphatidylinositol (PI) hydrolysis, while group II and HI mGluRs are negatively coupled to cyclic adensoine-3', 5'-monophosphate (CAMP) production. Opioid receptors are also coupled to these same systems, and have been shown to elicit changes in these messenger systems during chronic treatrnent.

We showed that chronic intracerebroventncular (i.c.v.) administration of selective group II and III mGluR antagonists concurrently with subcutaneous (SC.) morphine significantly reduced the severity of precipitated withdrawal symptoms.

Conversely, acute i.c.v. injection of a selective group II mGluR antagonist just pnor to the precipitation of withdrawal significantly exacerbated the severity of abstinence symptoms. In addition, acute icv. injection of a selective group II mGluR agonifi just pnor to the precipitation of withdrawal significantly reduced abstinence symptoms. From these results we hypothesized that chronic opioid treatment may induce a desensituation of group II rnGluRs.

We also demonstrated that chronic i.c.v. infusion of a selective group 1 mGluR antagonist concurrently with sac.morphine significantly attenuated the precipitated withdrawal syndrome. In addition, we showed that chronic i.c.v. antagonism of 6- opioid receptors with a highly selective antagonist dso decreased the development of morphine dependence, as well as tolerance. Since both group 1 mGluRs and 3-opioid receptors are positively coupled to PI hydrolysis, fùrther evidence for a role of products of PI hydrolysis in the development of morphine dependence was obtained when we showed that selective chronic inhibition of protein kinase C (PKC) activation, as weii as selective chronic inhibition of intracellular ca2+release, concurrently with morphine treatrnent significantly reduced the severity of abstinence symptoms. Thus, compensatory changes usuaily elicited by chronic opioid treatment may be counteracted by antagonizing receptors positively coupled to PI hydrolysis, as well as by inhibiting products of PI hydrolysis.

In the General Discussion, we propose a mode1 based on the possible interaction of mGluRs and opioid receptors, via related intraceliular second messengers, to explain the development of morphine dependence. Résumé

Nous avons étudié le rôle des récepteurs métabotropiques du glutamate

(mGluRs) et des seconds messagers intracellulaires associés dans le développement de la tolérance et dépendence à morphine. Les mGluRs sont divisés en trois groupes: les mGluRs du groupe I sont couplés positivement a l'hydrolyse du phosphatidylinositol

(PI), alors que les mGluRs des groupes II et III sont couplés négativement à la production d'adenosine 3',5'-monophosphate cyclique (CAMP). Les récepteurs aux opiacés sont également couplés aux mêmes systèmes et on a montré qu'ils produisent des changements des seconds messagers lors des traitements chroniques.

Nous avons montré que l'administration intracérébroventriculaire (i.c.v.) d'antagonistes sélectifs des mGluRs des groupes II et III concurrement à celle de morphine en sous cutané (s.c.) réduit de manière significative la sévérité des symptômes de retrait provoqué. Inversement, l'injection i.c.v. aiguë d'un antagoniste des mGluRs des groupes II juste avant la provocation de retrait augmentait de façon significative la sévérité des symptômes d'abstinence. De plus, l'injection i.c.v. aiguë d'un agoniste sélectif des mGluRs de groupe U just avant la provocation de retrait

élucidait de manière significative les symptômes d'abstinence. Ces résultats nous conduisait à faire l'hypothèse qu'un traitement chronique aux opiacés pourrait produire une désensibilisation des mGluR du groupe II.

Nous avons aussi démontré que l'infusion chronique i.c.v. d'un antagoniste sélectif des mGluRs du groupe 1, concurrernent avec la morphine s-c.,atténue de façon sélective le syndrome de retrait provoqué. De plus, nous avons montré que l'antagonisation chronique i.c.v. des récepteurs 6 aux opiacés à l'aide d'un antagoniste hautement sélectif réduit le developpement de la dépendence aussi que de la tolérance

à la morphine. Comme à la fois les mGluRs du groupe 1 et les récepteurs 6 aux opiacés sont couplés positivement à l'hydrolyse de PI, une évidence supplémentaire du rôle des produits de I'hydrolyse de PI dans la dépendence à la morphine a été obtenue lorsque nous avons montré que l'inhibition sélective chronique de l'activation des protéines kinases C (PKC), tout comme I 'inhibition sélective chronique de la libération intracellulaire de ca2', concurrement à un traitement par la morphine réduit significativement la sévérité des symptômes d'abstinence. Ainsi, les changements compensatifs produits par un traitement chronique par les opiacés peuvent être neutralisés en antagonisant les récepteurs qui sont couplés positivement à I'hydrolyse de PI ou bien en inhibitant les produits de l'hydrolyse de PI.

Dans la discussion générale, nous proposons un modèle basé sur les interactions possibles entre les mGluRs et les récepteurs aux opiacés, via seconds messagers intracellulaires associés, pour expliquer le développement de la dépendance a la morphine. CHAPTER 1: INTRODUCTION 1. Brief overview of opioids

The sensation of pain is the body's way of indicating the presence or threat of potential tissue injury. As injury may lead to debilitation or death, the sensation of pain cm be considered to be an efficient suMvai mechanism in most situations.

However, there are times when the sensation of pain provides little useful information or is itself debilitating, such as in chronic pain or post-surgical pain. In these situations, it is desirable to minimize pain. Analgesic dmgs are useful in this respect.

In particular, opioid dmgs such as morphine are widely used in the management of pain. Naturaily occurring opioids are derived from the opium poppy, Pcrpmer

Somni$emm. The juice from this poppy, opium, has been used since at least the third century B.C., and the psychological effects may have been known to the ancient

Sumerians at the end of the third millenium B.C. Initial use of opium may have been for spiritual purposes or its euphonc effect, and only later was it used in medicine to relieve pain and as an anaesthetic for surgery. In 1806, Serturner isolated morphine from opium. The narne morphine came fiom the Greek god of dreams, Morpheus.

Soon after, other alkaloids were discovered, such as codeine by Robiquet in 1832, and papaverine by Merck in 1848. Although synthetic opioids have been rnanufactured, morphine is considered to be the prototypic opioid dmg, and is still the most commonly used anaigesic (Brownstein, 1993; Jaffe and Martin, 1990).

Morphine acts primanly at p-opioid receptors. There are also opioid receptors termed 6- and K-opioid receptors, as well as E-opioid receptors and the more controversiai o-opioid receptors. Analgesia can be elicited by activating opioid receptors with various agonists. Long-term use of opioid dmgs, however, leads to the development of tolerance and dependence. In the first part of this introductory chapter, I will discuss the discovery of opioid receptors and endogenous ligands. opioid receptor subtypes, and the characteristics of opioid tolerance and dependence.

Since recent evidence indicates that activation of excitatory amino acid (EAA) receptors may be critically involved in the development of morphine tolerance and dependence, the second part of the introduction will ded with EAA receptors and their involvement in the development of opioid tolerance and dependence. In the third and final section of the introduction, I wiIl review the effects of acute and chronic administration of opioids on cyclic adenosine-3',St-monophosphate(CAMP) production and phosphatidylinositol (PI) hydrolysis, as activity of these intracellular second messenger systems has been show to be altered by opioid receptor activation.

A. Opioid receptors and endogenous ligands

Because morphine is such an effective analgesic, it was hypothesized that there must be receptors in the central nervous system which would bind morphine. In the early 19701s, stereospecific receptor sites for morphine were isolated in brain and guinea pig ileum, and their distributions were characterized (Kosterlitz, Lord and

Watt. 1972; Kuhar, Pert and Snyder, 1973; Hiller, Pearson and Simon, 1973; Terenius,

1973).

The existence of specific opioid receptors in the central nervous system suggested that there could be endogenous opioid ligands. In 1975, Hughes isolated a morphine-like peptide from rabbit, guinea pig, rat and pig brains. He found the highest concentrations in the stnatum, rnidbrain, pons and medulla. At this time the

"morphine-like" compound was not given a narne. Later that same year, Hughes and colleagues characterized the structure of two endogenous opioids from pig brain: H-

Tyr-Gly-Gly-Phe-Met-OH (Met-enkephalin) and H-Tyr-Gly-Gly-Phe-Leu-OH (Leu- enkephalin) (Hughes, Smith, Kosterlitz, Fothergdl, Morgan and Moms, 1975).

Shortly after, the same peptides were isolated from bovine brain (Simantov and

Snyder, 1976). Isolation of the other endogenous opioids, endorphin (Bradbury,

Srnyth and Snell. 1976; Li and Chung, 1976; Ling, Burgus and Guillemin, 1976) and dynorphin (Goldstein, Fischi, Lowney, Hunkapiller and Hood, 198 1) followed shonly.

Deltorphins were also isolated fiom the skin of the frog Phyllomeairsa bicolor

(Brownstein, 1993; Erspamer, Melchiom, Falconeri-Erspamer, Hegri, Corsi, Severini.

Barra, Sirnrnaco and Kriel, 1989).

The opioid peptides are derived from larger proteins called precursor proteins.

Proenkephalin contains the sequences for Met-enkephaiins, Leu-enkephalin, Met- enkephalin-kg6-phe7, and ~et-ertkephalin-~r~~-~l~~-~eu*(Elrownstein, 1993 ; Noda,

Furutani, Takahashi, Toyosato, Hirose, Inayarna, Nakanishi and Numa, 1982).

Dynorphin A, dynorphin B, a-neoendorphin and P-neoendorphin are al1 derived from prodynorphin (Brownstein, 1993; Kakidani, Funitani, Takahashi, Noda, Monmoto,

Hirose, Asai, Inayarna, Nakanishi and Numa, 1982). P-endorphin, corticotropin and a-melanotropin are derived fiom pro-opiomelanocortin (Brownstein, 1993; Nakanishi, houe, Kita, Nakarnura, Chang, Cohen and Numa, 1979). Because not al1 opioids have identical effeas, it was suspected that different opioid receptors existed. Chang and Cuatrecasas ( 1979) demonstrated, using binding cornpetition studies, that morphine and enkephdins bind to different receptors.

Morphine receptors were termed mu (p) receptors and enkephalin receptors were termed delta (6) receptors. Morphiceptin and derrnorphin are thought to be endogenous ligands for the p receptor (Wollemann, Benyhe and Simon, 1993).

Martin, Eades, Thompson, Hupler and Gilbert (1976) fist described kappa (K)and sigma (o)opioid receptors. Endogenous substances thought to be selective for K receptors are the dynorphins and ~et-enke~halin-~r~~-~he~(Brownstein, 1993 ;

Wollema~et al, 1993). The a receptor is no longer considered to be an opioid receptor because the non-selective opioid antagonist naloxone has no action on it

(Wollemann et al, 1993). Furthemore, Li and Chung (1976) characterized the epsilon

(E) receptor, at which the endorphins are believed to be the endogenous ligands.

Because most studies of opioid action are perfomed using either morphine or synthetic opioids selective to p-, 6- or K-opioid receptors, the E-opioid receptor and endorphins are not a focus of this paper.

Opioid receptors are seven-transmembrane guanine nucleotide-binding (G) protein-coupled receptors which mediate effects on CAMP production, PI hydrolysis, and ion channel activity. Recently, p-, 6- and K-opioid receptors have been cloned.

The first opioid receptor to be cloned was the 6 receptor with cDNA isolated from

NG108-15 cells (Evans, Keith, Morrison, Magendzo and Edwards, 1992; Kieffer,

Befort, Gaveriaux-Ruff and Hirth, 1992), mouse brain (Yasuda, Raynor, Kong, Breder. Takeda, Reisine and Bell, 1993) and rat brain tissues (Fukuda, Kato, Mori,

Nishi and Takeshima, 1993). Yasuda et al (1993) also isolated a mouse brain K-opioid receptor clone. Chen, Mestek, Liu, Hurley and Yu (1993) cloned the p-opioid receptor from a rat brain cDNA library. The amino acid sequences of the membrane spanning segments 2, 3 and 7, as weii as the intraceliular loops, are highly conserved between cloned opioid receptors (Reisine and Bell, 1993). The second and third intracellular loops may be the regions interacting with G-proteins (Reisine and Bell,

1993). The amino and carboxyl termini, as well as the second and third extracellular loops, are quite different among the various cloned opioid receptors, suggesting that these regions may be involved in ligand binding (Reisine and Bell, 1993). The phmacology of cloned opioid receptors is similar to that of opioid receptors in the central nervous system (Reisine and Bell, 1993).

High levels of mRNA expression for the p-opioid receptor have been found in thalamus, stnatum, locus coeruleus and nucleus of solitary tract (Mansour. Fox, Akil and Watson ( 1995). High ievels of 6-opioid receptor mRNA expression are found in stnatum and cortex (Mansour et al, 1995), while K-opioid receptor mRNA is highly expressed in nucleus of solitary tract, hypothalamus, nucleus accumbens, substantia nigra and ventral tegmental area (Mansour et al, 1995).

B. Opioid receptor subtypes

In the last decade, it has been hypothesized that p, 6 and K receptors are further divided into subtypes. For example, p receptors are divided into two subtypes: p,-receptors and y-receptors. The p-opioid receptors have been classified into pi, to which morphine and most enkephahs bind with high afhity, and ~q-receptors,which preferentially bind morphine (Goodman and Pasternak, 1985; Hazum, Chang,

Cuatrecasas and Pastemak, 198 1; Moskowitz and Goodman, 1985; Nishimura, Recht and Pasternak, 1984; Paul, Bodnar, Gistrak and Pastemak, 2989; Pasternak and

Wood, 1986; Wolozin and Pastemak, 198 1; Zhang and Pastemak, 1980). It has been postulated that pl receptors are implicated in supraspinal analgesia (Heyman, Williams.

Burks, Mosberg and Porreca, 1988; Paul et al, 1989), while pz receptors are implicated in spinal analgesia (Paul et al, 1989). It is proposed that the 61 receptor subtype is sensitive to the agonist [I3-pen2, D-~en']enke~halin(DPDPE) and the antagonist [D-Ala2. eu^, cys6]enkephalin@ALCE), whereas the & receptor subtype is sensitive to the agonist @3-AJa2, ~lu"]deltor~hin(DELT II) and the antagonist naltrindole-5'-isothiocyanate (5'-NTII) (Horan, Wild, Misicka, Lipkowski, Haaseth,

Hruby, Weber, Davis, Yamarnura and Porreca, 1993). Both 6 receptor subtypes are believed to be involved in nociception at the spinal level (Sofuoglu, Ponoghese and

Takemori, 1993). The K receptor subtypes have been classified as KI,which preferentially binds arylacetamides such as U50,488 and U50,69,593 (Horan rf al,

1993; Wolleman, et al, 1993), and KZ, which preferentially binds benzomorphans such as bremazocine and ethylketocyclazocine (EKC) (Horan ef al, 1993). Currently cloned receptors correspond to the pl-, 62- and ~~-opioidreceptor subtypes.

Moreover, it has been hypothesized that p receptors may exist both independently, and functionally coupled to 6 receptors (Rothman and Westfall, 1982;

Tiseo and Yaksh, 1993; Vaught, Rothman and Westfall, 1982). Some of the first evidence suggesting the possible existence ofa p/6 complex came from studies

examining the effects of the endogenous 6 ligands, leu-enkephalin and met-enkephalin,

on morphine-induced analgesia. It was observed that low anaigesic doses of leu-

enkephalin potentiated morphine-induced anaigesia (Larson, Vaught and Takemori,

1979; Lee, Leybin, Chang and Loh, 1980; Vaught and Takernori, 1979; Vaught et al,

1982). while met-enkephalin either had no effect (Vaught and Takemori. 1979) or

inhibited analgesia induced by high doses of morphine (Lee et al, 1980; Vaught et al,

1982). Because binding studies indicated that morphine and the enkephaiins preferred

different receptors, it was hypothesized that morphine and enkephalin receptors

existed in a receptor complex whereby activation of enkephalin receptors could

modulate the fùnctioning of morphine receptors. In this context, leu-enkephalin was

considered an "agonist" whereas met-enkephaiin was considered an "antagonist"

(Vaught et al, 1982). Vaught et al (1982) described a hypotheticai model for the jd6

receptor complex. The p receptor is directly coupled to an effector mechanism, and

binding of an agonist to this receptor induces a conformationai change which

stimulates the effector mechanism, and elicits analgesia. According to this model, the

6 receptor is allosterically coupled to the p receptor, and/or directly coupled to the

effector mechanism. Binding of an agonist, such as leu-enkephalin, to the 6 receptor could induce a confornational change which potentiates the stimulation of the effector mechanism by a p. agonist. Binding of an antagonist, such as met-enkephaiin, would either not potentiate p-receptor activation of the effector mechanism, or induce a conformational change which inhibits activation of the effector mechanism. There may also be ligands which bind to both the p and 6 receptors in the complex simultaneously. Examples of these "self-potentiating" ligands may be P- endorphin and DADLE. It is generally agreed that analgesia to noxious heat stimuli is mediated pnmarily via p receptors (Vaught et al, 1982), and although DADLE oniy has 1/7 the affinity for the p receptor as morphine, and P-endorphin has oniy a slightiy greater affinity, both compounds are many times more potent in the mouse tail-flick test. It is hypothesized that the binding of these "self-potentiating" compounds to the

6 receptor potentiates their actions at the p receptor (Vaught eî al, 1982). Blocking the 6 effects of these self-potentiating agents with met-enkephdin greatly reduces their analgesic efficacy .

Further evidence for an allosteric p/6 coupling is that the selective 6 antagonist naltrindole (Portoghese, Sultana and Takemori, 1988) can antagonize morphine anaigesia to some extent (Tiseo and Yaksh, 1993). Based on the data of Tiseo and

Yaksh (1 993), who found that naltrindole partially blocked morphine-induced analgesia, and the suggestion of Stewart and Hammond (1993) that naltrindole prefers the 6, receptor, it could be hypothesized that the 6, receptor is part of the p/6 cornplex described earlier. Stewart and Hammond (1993) also found that the selective 6, receptor antagonist naltrindole attenuates the effect of the selective p agonist

DAMGO in the hot-plate test in rats. Moreover, it has also been demonstrated that a selective y antagonist, CTOP,can antagonize analgesia induced by DADLE (Suh and

Tseng, 1990), a putative 6, agonist. However, other investigators have found that naltrindole does not affect morphine-induced analgesia (Calcagnetti and Holtzman, 1991). It is also important to note that there is evidence suggesting that naltrindole may have sigmficant antagonist activity at p receptors (Portoghese,

Nagase, MaloneyHuss, Lin and Takemon, 199 1).

C Tderance und dependence

Although opioid drugs such as morphine are widely used for the management of pain, their clinical usefulness is limited by the development of tolerance and dependence with their chronic use. Tolerance is a decreased sensitivity to the effect of a drug following its repeated or continuous adrnnistration. In the case of opioids. analgesic tolerance is of great importance. A decrease in the analgesic efficacy of an opioid dmg may be demonstrated in two ways: 1) a progressively lower level of analgesia produced by a given treatment dose on a pain test (Adams and Holtzman,

1990). or 2) a nghtward shift and flattening of the log-dose response (LDR) cuve

(Adams and Holtzman, 1990; Mucha, Kalant and Linseman, 1979; Mucha and KaIant,

1980; Stevens and Yaksh, 1989qb). A nghtward shift of the LDR curve is indicative of decreased potency of the drug, necessitating the use of a higher dose to elicit effects previously produced by lower doses of the drug. A flattening of the curve is indicative of a decrease in the maximum effect which can be attained (Mucha and Kalant, 1980).

Higher treatment doses and longer durations of treatment produce greater changes in the LDR curve, and a flattening of the curve may not be evident until very high treatment doses are used (Mucha and Kalant, 1980).

Dependence is a continued need for the dmg, after chronic use, to maintain a state of physiological equilibriurn, and resuits in an aversive withdrawal syndrome upon rernoval of the drug. The drug can be removed by simply discontinuing administration, however, precipitated withdrawal is more commonly studied experirnentally. To precipitate withdrawal, an opioid antagonist such as naloxone is injected to competitively block the opioid receptors (Etuckett, 1964; Cochih 1973;

Kaymakcalan and Woods, 1956; Lemberger and Rubin, 1976). Mer simple discontinuation of drug treatment, symptoms are slow to appear and rnay take days to reach their peak, the onset being related to the clearance rate of the drug. Injection of an antagonist blocks opioid receptors rapidly, terminating the actions of the agonist, and symptoms are generally severe and appear within minutes. The degree of dependence is quantified by the severity of symptoms dunng the withdrawal period.

The development and expression of tolerance and dependence has been described for many species including humans, dogs, cats, rabbits, rats and mice

(Halbach and Eddy, 1963; Tatum, Seevers and Collins; 1929). In humans, withdrawal symptoms resemble a bad case of influenza and may include yawning, lacrimation, rhinorrhea, perspiration, mydriasis, tremor, gooseflesh, anorexia, restlessness, emesis, fever, hyperpnea, increased systolic blood pressure, weight loss and stomach cramps

(Halbach and Eddy, 1963). Most current experimentai studies use rodents as subjects.

In rats, symptorns include teeth chattering, jumping, wet dog shakes, writhing, ptosis, lacnmation, eye twitch, salivation, rhinorrhea, diarrhea, screarn on touch, hostility on handling, and penile erection (Blasig, Herz, Reinhold and Zieglgansberger, 1973;

Halbach and Eddy, 1963). In mice, investigators usuaily only measure jumping as other behaviours are difficult to quanti& in these animals. Morphine tolerance and dependence have been demonstrated with systernic, spinai and supraspinal routes of

administration (Cowan, Zhu, Mosberg, Omnaas and Porreca, 1 988; Fundytus, 1992;

Stevens and Yaksh, 1989; Yobum, Lutfy, Sierra and Tortella, 1990).

i) Relationship Between u and 6 Rece~tors

Morphine is the prototypical opioid upon which studies of tolerance and dependence have been based. Morphine is primarily a p agonist, with some effects at

6 and K receptors (Takemon and Portoghese, 1987). As discussed previously, some

investigators hypothesize the existence of a pl6 opioid receptor complex (Rothrnan and Westfall, 1982; Vaught et aI,1982). Thus, the questions of whether or not tolerance and dependence develops to 6 agonists, whether there is cross-tolerance between p and 6 agonists, and whether 6 ligands cm affect the development of tolerance and dependence to chronically administered morphine have been exarnined.

Until recently, no 6 agonists which are systernically active have been available.

Therefore, studies examining the relationship between p and 6 receptors have generally been performed using spinal and central administration.

It has been shown that tolerance develops to the analgesic effects of spinally and supraspinally administered 6 agonists (Stevens and Yaksh, 1989a,b; Suh and

Tseng, 1990; Kalso, Sullivan, McQuay, Dickenson and Roques, 1993; Kovacs,

Nyolczas, Krivan and Gulya, 1988) at a rate similar to that of p agonists. Stevens and

Yaksh (1 989% b) noted that dependence dso developed in rats treated intrathecally with 6 agonists. Moreover, they concluded that the 6 agonist DADLE induced more dependence than y agonists, as evidenced by more wet dog shakes during the withdrawal period. However, the only syrnptorn they measured was wet dog shakes.

and wet dogs shakes are considered to be a recessive abstinence symptom (Blasig et

al, 1973). Recessive abstinence symptoms occur more frequently when withdrawal is

less severe. When withdrawal is more severe, symptoms such as teeth chattering and jumping preclude the occurrence of wet dog shakes. Therefore, although they

observed more wet dog shakes in rats treated with 6 agonists, without measures of

other symptoms, these results cannot be interpreted conclusively. Most investigators

have found that cross-tolerance between selective p and 6 agonists does not occur

(Suh and Tseng, 1990qb; Desmeules, Kayser, Gacel, Guilbard and Roques. 1993;

Kalso et al, 1993; Porreca, Heyman, Mosberg, Omnaas and Vaught, 1987). However,

some investigators have found unidirectionai cross-tolerance between p and 6

agonists. Yobum et al (1993) observed that rnice tolerant to the analgesic effects of

subcutaneous morphine in the tail-flick test were also tolerant to the 6 agonist DPDPE

administered intrathecally (i.t.) and intracerebroventricularly (icv.), but not vice-

versa.

It has been hypothesized that treatment with 6 ligands may decrease the

seventy of the morphine withdrawal syndrome. Blockade of 6 receptors with the

selective antagonists naltrindole, naltrindole-5'-isothiocyanate(5'-NTII) and naltriben

has been shown to attenuate the development of tolerance and the seventy of the

precipitated withdrawal syndrome in mice (Abdelhamid, Sultana, Portoghese and

Takemori, 199 1; Miyamoto, Portoghese and Takemon, 1993). Interestingly,

concurrent treatment with a selective and non-peptidergic 6 agonist, BW373U86, was also shown to attenuate the severity of the precipitated withdrawal syndrome (Lee,

McNutt and Chang, 1993). This effect was antagonized by naltrindole. Thus, 6 receptor activation appears to be involved in the development of morphine tolerance and dependence. ii) Relationship between u. and k receptors

Several investigators have also been interested in whether tolerance and dependence develops with chronic use of K agonists, whether there is cross-tolerance between p and K agonists, and whether K ligands can modulate the developrnent of tolerance and dependence to morphine.

First, it has been demonstrated that tolerance and dependence will develop when K agonists are administered chronically. The developrnent of analgesic tolerance and physical dependence to the selective K agonists U50,488H and Mr2033 have been demonstrated in both rats and monkeys (Bhargava, Gulati and Rarnarao, 1989a;

Bhargava, Ramarao and Gulati, 1989b; Grnerek, Dykstra and Woods. 1987: Young and Khazan, 1985). Although cross-tolerance between specific K agonists such as

U50.488H and p agonists does not usually occur (Bhargava el al, 1989b; Gmerek et al, 1987), cross-tolerance between p agonists and less specific K agonists such

Mr2033 and ethylketocyclazocine (EKC) has been demonstrated (Gmerek et al, 1987;

Hong, Young and Khazan, 1986). However, these less specific r agonists are believed to have some activity at p receptors, as well as their primary activity at K receptors

(Horan et al, 1993). Furthemore, some investigators have found that IC antagonists cm precipitate withdrawal in morphine treated rats, although the symptoms are less severe than those precipitated by p antagonists (Maldonado, Negus and Koob, 1992;

Ramabadran, 1985). These results suggest that aithough p receptors play a primary role in morphine tolerance and dependence, K receptors may also be involved.

Another question which has recently been addressed is whether concurrent administration of K ligands with morphine treatment modulates the development of morphine tolerance and dependence. Several investigators have found that concurrent treatment with the K agonists dynorphin(1-13), dynorphin(1-17) and U5û,488H attenuated the development of morphine tolerance and dependence in mice and rats

(Takemori, Loh and Lee, 1993; Yamamoto, Ohno and Ueki, 1988). However, other investigators have observed no effect of concurrent K agonist treatment on the severity of precipitated morphine withdrawal (Fukagawa, Katz and Suniki, 1989). Therefore, there is some uncertainty as to the role activation of K receptors plays in the development of morphine tolerance and dependence. It is not known whether concurrent treatment with selective K antagonists can modulate the development of morphine tolerance and dependence.

An interesting analgesic agent is butorphanol, which is believed to act at p, 6 and K receptors. It has been demonstrated that tolerance and dependence do develop with chronic administration of butorphanoi (Jaw, Makimura, Hoskins and Ho, 1993b;

Jaw, Hoskins and Ho, 1993a). Concurrent treatment of rats with butorphanol and either p, 6 or K antagonists decreased the severity of the precipitated withdrawal syndrome (Jaw et al, 1993a; Jaw et al, 1993b; Oh, Makimura, Jaw, Hoskins and Ho, 1992). These results provide further evidence for involvement of 6 and K, as well as p, receptors in the development of opioid tolerance and dependence.

CI. Excitatory amino acids

Since the 19303, the excitatory amino acid glutamate (and aspartate) and its actions in the brain has been extensively studied. It has been suggested that early studies conducted by Krebs (1935) pointed to a central metabolic role for glutamate in the brain. There were claims that glutamate was helpfùl in neurological disorders such as epilepsy and mental retardation, and it was recognized that glutamate metabolism in the brain is very complex (Fomum, 1984; Berl, Lajtha and Waelsch, i 96 1).

Electrophysiological studies showed that glutamate has excitatory effects on spinal cord neurons (Curtis and Watkins, 196 1). Furthermore, glutamate satisfies the four main criteria to be classed as a neurotransrnitter: 1) it is pre-synaptically localized in specific neurons; 2) it is specifically released by physiological stimuli in concentrations high enough to elicit post-synaptic response; 3) it demonstrates identity of action with the naturally occumng transrnitter, including response to antagonists; and 4) mechanisms exist that will terminate transmitter action rapidly (Fomum, 1984).

More recently, the role of glutamate in neuronal plasticity and neurotoxicity has been recognized. Learning and memory, involving long-term potentiation (LTP). has been shown to involve activation of glutamate receptors (Collingridge and Lodge,

1987). In addition, neurodegenerative disorders, such as epilepsy and stroke, have been shown to involve excessive glutamate release and glutamate toxicity (Nicoletti,

Bruno. Copani, Casabona and Knopfel, 1996). A. Receptors

The endogenous excitatory amino acids (EAAs) glutamate and aspartate activate several different types of recepton. There are currently believed to be three families of ionotropic receptors (NMDA, AMPA and kainate) and a family of metabotropic receptors at which they act. i) NMDA receptors

The NMDA EAA receptor subtype was so named because of the selective binding of N-methyl-D-aspartate VA)(Lodge, 1992) at this site. A high concentration of NMDA receptors can be found in hippocampus, cortical structures. basal ganglia and sensory-associated systems (Cotrnan and Monaghan, 1988). as well as the spinal cord.

NMDA receptors are known as ionotropic EAA receptors because activation of these receptors leads to opening of its associated ion channel. Activation of NMDA receptors leads to influx of ca2+and Na' (with a concurrent efflux of K*)

(MacDermott, Mayer, Westbrook, Smith and Barker, 1986; Mayer, MacDermot.

Westbrook, Smith and Barker, 1987) through the NMDA channel. Influx of ca2* following stimulation of NMDA receptors leads to activation of several intracellular second messenger systems. Furthermore, NMDA receptors are gated in a voltage- dependent rnanner by M~'' (Nowak, Bregestovski, Ascher, Herbet and Prochiantz,

1984; Mayer, Westbrook and Guthrie, 1984). It has been shown that the NMDA response in neuronal cultures is enhanced in ~~~'-freesolutions, and that the addition of M~~'to the culture medium blocks inward current flow through NMDA receptors (Nowak et al, 1984; Mayer, Westbrook and Guthrie, 1984). lt should be noted that in physiological concentrations of M~*',the block is voltage-dependent and occurs at potentials more negative than -20 mV (Cotman and Monaghan, 1988). which includes the normal resting potentids of neurons. It has been proposed that M~~'binds to a site within the channel to produce the block.

It is generally believed that the NMDA receptor has several components that interact allostencally (Cotman and Monaghan, 1988). There are several binding sites on NMDA receptors including an agonist (NMDNgJutamate) site, an allostenc glycine site, and a phencyclidine (PCP) site (Cotman and Monaghan, 1988). Binding at the glycine site potentiates the response of the receptor to agonist activation, whereas binding at the PCP site produces a non-cornpetitive antagonism of the neuron to agonist stimulation.

Agonists of the NMDA receptor acting at the glutamate binding site include glutamate, aspartate, ibotenate and, of course, NMDA. Glycine is an agonist at the allostenc glycine binding site and enhances the response of the NMDA receptor to stimulation at the glutamate binding site. Substances which bind to the PCP site (such as PCP and MK-80 1 (+- 10,ll-dihydro-5-H-dibenzo-cycloheptene)) are non- cornpetitive antagonists and inhibit the response of the NMDA receptor to stimulation at the glutamate binding site (Wong, Kemp, Priestley, KNght, Woodruff and Iversen,

1986). Note that it is necessary to have a minimal amount of activity at the glutamate binding site before these non-cornpetitive antagonists can bind to the PCP site and produce their effect. For this reason they are referred to as use-dependent and have been labelled "open-channel" blockers. These non-cornpetitive antagonists inhibit activaion of NMDA receptors by preventing the influx of caz'. Substances which prevent activation of the NMDA receptor by competing for binding at the glutamate binding site are known as cornpetitive antagonists. Examples of this type of antagonist are D-APV @-2-aminophosphonovalerk acid) and CPP (3-(2-carboxypiperazine-4- yl)propyl- l -phosphonic acid).

Recently, NMDA receptors, including several splice variants, have been cloned

(Nakanishi, 1994). There are currently believed to be five NMDA receptor subunits:

NMDAR 1. NMDAIUA to 2D (Nakanishi, 1994; Moriyoshi, Masu, Ishiik, Shigemoto,

Minino and Nakanishi, 199 1; Masu, Tanabe, Tsuchida, Shigemoto and Nakanashi,

199 1; Ishii. Monyoshi, Sugihara, Sakurada, Kadotani, Yokoi, Akazawa, Shigemoto.

Minino, Masu and Nakanashi, 1993; Meguro, Mon, Araki, Kushiya, Kutsuwada,

Yarnazaki, Kumanishi, Arakawa, Sakimura and Mishini, 1992; Kutsuwada,

Kashiwabuchi, Mon, Sakimura, Kushiya, Araki, Meguro, Masaki, Kimanishi, Arakawa and Mishina, 1992; Monyer, Sprengel, Schoepfer, Herb, Higuchi, Lomeli, Bumashev,

Sakmann and Seeburg, 1992). The NMDA receptor subunits share sirnilar structural characteristics, including four transmembrane segments, and are sirnilar to other ligand-gated ion channels (Nakanishi, 1994). NMDA receptors have much larger amino terminal portions than other ligand-gated ion channels, and NMDAR2 subunits also have a large extracellular C-terminal extension (Nakanishi, 1994). Although cells in which NMDAR.2 subunits are expressed alone show no electrophysiological currents derglutamate stimulation, expression of the NMDAR2 subunit in combination with NMDARl potentiates the activity of MAR1(Nakanishi, 1994). ii) AMPA Rece~tors

Initially called "quisqualate" receptors. AMPA receptors are fast acting receptors (Cull-Candy and Ogden, 1985; Kiskin, Krishtal and Tsyndrenko, 1986) whjch gate voltage-independent ion channels. They gate cation channels which are permeable to Na-, but for the most part are not permeable to ca2*(Murphy and Miller.

1989). Continuous activation of AMPA receptors depolarizes the cell, relieving

NMDA receptors of the voltage-dependent M~~',block and aiding in activation of

NMD A receptors. Agonists at AMP A receptors include glutamate, aspartate, quisqualate and, as the narne indicates, AMPA (a-amino-3-hydroxy-5-methyl-4- isoxazole propionic acid).

Four AMPA receptors have recently been cloned. The iGluR subunits termed

GluR I to GluR4 (or GluRA to GluRD) bind AMPA with a higher affinity than kainate

(Bettler, Boulter, Hermans-Borgmeyer, O'Shea-Greenfield, Deneris, Moll,

Borgmeyer, Hoiimam and Heinemann, 1990; Gasic and Hollemann, 1992; Hollmann,

O-Shea-Greenfield, rogers and Heinemann, 1989; Keinanen, Wisden, Sommer,

Werner, Herb, Verdoom, Sakmann and Seeburg, 1990). These receptors have four transmembrane spanning domains, sirnilar to other ligand-gated ion channeis (Gasic and Hollemann, 1992). In neurons expressing GluR.2 with other subunits, there are no ca2*currents, however if the GluR2 subunit is not present, glutamate cm trigger ca2' influx (Gasic and Hollmann, 1 992; Miller, 199 1 ; Sommer and Seeburg, 1992). iii) Kainate receptors

As with AMPA receptors, kainate receptors gate cation channels that are petmeable to Na', but have negligible permeability to ca2' (Murphy and Miller, 1989).

However, as previously mentioned, there are AMPAkainate receptors that show ~a'*

(Miller, 199 1; Sommer and Seeburg, 1992). Similarly to AMPA receptors, kainate receptors are involved in fast excitatory responses to EAA transmitters (Cd-Candy and Ogden, 1985; Kiskin et al, 1986), relieving the NMDA receptor of its voltage- dependent M~~*block.

Substances which activate kainate receptors include glutamate, aspartate, and kainate. Although distinct families of high-afinity AMPA and kainate receptors have been isolated, the fùnctional distinction between these receptor subtypes is not clear

(Barnard and Henley, 1990).

Ionotropic glutamate receptors which bind kainate, but show no affinity for

AMPA, have recently been cloned (Gasic and Hollmann, 1992; Bettler et al, 1990;

Bettler, Egebjerg, Shma, Pecht, Hermans-Borgmeyer, Moll, Stevens and

Heinemann, 1992; Egebjerg, Bettler, Hermans-Borgmeyer and Heinemann, 199 1).

The GIuR5 - 7 subunits are kainate selective (Bettler et al, 1990; Bettler et al, 1992;

Egebjerg et al, 199 1). As with AMPA receptors, GluR5 to 7 are proposed to have four transmembrane domains, similar to other ligand-gated ion channels (Gasic and

Hollrnann, 1992). There is also a family of kainate binding proteins (e.g. KA-1, KBP- f. KBP-c) which bind kainate with high atfinity, but are not fundional receptor channels and have little sequence similarity with GluR genes (Gasic and Hollmann,

1992). iv) Metabotropic receptors

The term "metabotropic glutamate receptor" or mGl@ was coined when it was observed that activation of certain quisqualate-sensitive receptors afTected intracellular second messenger systems. Subsequently, it was discovered that rnetabo tro pic glutamate receptors are coupled directly to the ce11 membrane b y a guanine nucleotide regulatory (G) protein (Sladeczek, Pin, Récasens, Bockaert and

Weiss, 1985; Sugiyama, Ito and Hirono, 1987) and are not associated with ion channels. Activation of mGluRs therefore produces direct effects on intracellular second messenger systems.

Several subtypes of mGluRs, fiom rnGluRl to mGluR8 have recently been cloned (Nakanishi, 1992; Okarnoto, Hori, Akazawa, Hayashi, Shigemoto, Mizuno and

Nakanishi, 1994; Saugstad, Kinzie, Mulvihill, Segerson and Westbrook. 1994;

Duvoisin et al, 1995; Zhang and Rarnonell, 1995). These cloned receptors can be classified into one of three groups based on receptor pharmacology, signaling cascades and sequence similanties (Hayashi, Sekiyama, Nakanishi, Jane, Sunter, Birse,

Udvarhelyi and Watkins, 1994). Group 1 mGluRs, consisting of rnGluR 1 and mGluR5, are positively coupled to PI hydrolysis. Activation of group 1 mGluRs stimulates phospholipase C (PLC), which catalyzes the hydrolysis of inositol-4,s- bisphosphate (Pl&) to inositol-1,4,5-tnsphosphate(IP,) and diacylglycerol @AG)

(Ambrosini and Meldolesi, 1989; Manzoni, Finiels-Marlier, Sassetti, Blockaert, Le Peuch and Sladeczek, 1990; Schoepp and Conn, 1993). IP3 stimulates the release of

ca2' fi-om intemal stores on the endoplasmic reticulum, and DAG promotes the

translocation and activation of protein kinase C (PKC) (Kapcala, Weng and Juang.

1992; Schoepp and COM, 1993). Group II rnGluRs, consisting of mGluR2 and 3, as

well group III mGluRs, consisting of mGluR4, 6, 7 and 8, are negatively coupled,

through a pertussis-toxin sensitive G-protein, to the production of CAMP (Hayashi ri

al. 1994). Although group II and III mGluRs have sirnilar effects on intracellular

second messengers, group III mGluRs show a difFerent receptor pharmacology in that

they are selectively activated by L-amino-4-phosphonobutanoate(L-AP4) (Hayashi et

ai, 1994; Saugstad et al, 1994), a receptor class which has been traditionally identified

as primarily presynaptic.

The general (Le. not selective to any specific subytpe) mGluR agonist I-

arninocyciopentane- l ,3-dicarboxylic acid (( I S, 3 R)-ACPD)activates al1 the mGluRs

and is ten times more potent as an agonist at mG1uR.s than ionotropic glutamate

receptors (Palmer, Monaghan and Cotman, 1989; Schoepp, Johnson, Salhoff.

McDonald and Johnson, 199 1; Schoepp, Johnson and MOM, 1992; Schoepp, Johnson.

True and Monn, 199 1 ; Watkins and Collingridge, 1994). Group I mGluRs are

selectively activated by (RS)-dihydroxyphenylglycine (DHPG) (Schoepp,

Goldsworthy, Johnson, Salhoff and Baker, 1994; Watkins and Collingridge, 1994).

Group II mGluRs are somewhat selectively activated by (1 S,3S)-ACPD and

2S, 1 'Et,2'R3'R)-2-(2',3'-dicarboxycyclopropyl)ycine @CG-IV) (Ishida, Saitoh,

Shimamoto, Ofine and Shinozaki, 1993; Pin and Duvoisin, 1995; Watkins and Collingridge, 1994). L-AP4 selectively activates group III mGIuRs (Foster and Fagg,

1984; Nakanishi, 1992; Nakajima, Iwakabe, Adawaza. Nawa, Shigemoto, Mimno and

Nakanishi, 1993; Okamoto, Hori, Akazawa, Hyashi, Shigemoto, ?wA5zunoand

Nakanishi, 1 994; Watkins and Collingridge, 1 994).

Group 1 mGIuR rnRNA is highiy expressed in cerebral cortex, cerebellum, stnatum, hippocampus and olfactory bulb (Masu et al, 1994). Group II mGluR mRNA is highiy expressed in cerebral cortex, cerebellum, striatum. hippocampus, thalamus. olfactory bulb and accessory olfactory bulb (Masu et al, 1994). There is a high level of expression for group UI mGluR mRNA in cerebellum, thalamus, olfactory bulb and retina (Masu et al, 1994). L-AP4 sensitive receptors have been show to be located pre-synaptically in brain and spinal cord (Koemer and Cotman,

198 1; Evans, Francis, Jones, Smith and Watkins, 1982; Davies and Watkins, 1982;

Monaghan, Bridges. and Cotman, 1989), and activation of these receptors ofien leads to a decrease in activity of post-synaptic receptors.

B. ficitatory nmino acids and morphine tolerance and dependence

Throughout the years, many neurotransmitter systems have been implicated in the development of tolerance to and dependence upon morphine. Recently, a role for the excitatory amino acid (EAA) system has also been suggested.

The relationship between opioids and excitatory amino acids becarne a topic of interest when it was observed that aspartate antagonized some morphine effects

(Koyuncuoglu, Güngor, Sagduyu and Eroglu, 1974) and attenuated the development of morphine tolerance and dependence (Koyuncuoglu, Güngor, Sagduyu and Eroglu, 1977). Furthemore, administration of morphine decreased levels of asparaginase, the enzyme responsible for catalyzing the synthesis of aspartate from asparagine

O

Güngor, Enginar, Hatipoglu and mal, 1986). Early studies showed that systernic administration of ketamine or dextromethorphan, NMDA antagonists, just prior to the precipitation of withdrawal decreased the seventy of abstinence symptoms

(Koyuncuoglu, Güngor, Sagduyu and Aricioglu, 1990). Conversely, pre-treatment with either NMDA antagonists or the opioid antagonist naloxone prior to any morphine treatment seemed to exacerbate some abstinence symptoms (Koyuncuoglu and Aricioglu, 199 1). It was hypothesized that opioids bound to EAA receptors and had an antagonistic effect, thereby eliciting supersensitivity dunng chronic opioid treatment (Koyuncuoglu et al, 1990; Koyuncuoglu and Aricioglu, 199 1), which would lead to over-activity when opioid agonist was removed from the receptor by an antagonist such as naioxone.

More recently. it has been found that concurrent treatment of rats with the non-selective EUantagonist kynurenic acid with daily injections of morphine was effective in attenuating the development of tolerance to morphine's analgesic effects

(Marek, B en-Eliyahu, Gold and Liebeskind, 199 1a). Similady, concurrent treatment of rats with the selective NMDA antagonist MK-801 with daily injections of morphine has also been shown to be effective in attenuating the development of tolerance to morphine's analgesic effects (Marek et al, 1991 a; Marek, Ben-Eliyahu, Vaccarino and

Liebeskind, 199 1b; Tmjillo and Akil, 1991,1994), as well as alleviating the seventy of some symptoms of the precipitated withdrawal syndrome (Trujillo and Akil,

I991,1994).

Some investigators have found treatment with the non-specific EAA antagonist kynurenic acid and the NMDA-specitic antagonist MK-801 ody on the day of testing

(Le. not concurrently with morphine) to be effective in decreasing the severity of' precipitated withdrawal symptorns (Koyuncuoglu, Dizdar, Aricioglu and Sayin, 1992;

Rasmussen, Fuller, Stockton, Peny, Swinford and Omstein, 199 1 ; Rasmussen, Krystal and Aghajanian, 199 1; Tanganelli, Antonelli, Morari, Bianchi and Beani, 199 1 ).

However, other investigators have found that this acute administration of MK-80 1 is ineffective at alleviating the severity of the precipitated withdrawal syndrome (Trujillo and Akil, 199 1).

Hyperactivity in the locus coeruleus (LC) has been shown to be correlated with the morphine withdrawal syndrome (Aghajanian, 1978; Valentino and Wehby, 1989).

Thus, the effect of EAA antagonists on LC hyperactivity during morphine withdrawal has also been investigated. Central administration of EAA antagonists into either the lateral ventncle or the locus coeruleus was found to decrease the hyperactivity of locus coeruleus neurons during precicpitated withdrawal (Akaoka and Aston-Jones, 199 1).

Thus, the non-selective EAA antagonist kynurenate, the selective competitive NMDA antagonist AP5, and the selective AMPAkainate antagonist CNQX reduced LC hyperactivity, with AP5 being the least effective. Conversely, systemic administration of either the non-cornpetitive NMDA antagonist MK-80 1 or the competitive NMDA antagonist LY2746 14 was unable to aEect the hyperactivity of LC neurons during

precipitated morphine withdrawal (Rasmussen et ai, 199 1).

The development of tolerance to and dependence upon morphine has been

discussed as a phenomenon of neuronal plasticity, perhaps resembling long-term

potentiation (Trujillo and Akil, 1991). NMDA receptors in the hippocarnpus have

long been hypothesized to be involved in the phenomenon of long-terni potentiation

(Collingridge and Singer, 1990). The effectiveness of selective NMDA receptor

antagonists in attenuating the development of tolerance to morphine's analgesic effects

and alleviating the seventy of the precipitated withdrawai syndrome lends support to

the hypothesis that the development of morphine tolerance and dependence is indeed a

fom of neuronal plasticity. It remains to be detennined just what neuronal changes

involving NMDA receptors, or EAA receptors in generai, are elicited by chronic

treatment with morphine. Although it has been suggested that supersensitivity of EAA

receptors is elicited by chronic opioid treatment (Koyuncuoglu et al, 1990), the fact

that chronic treatment with Ek4antagonists attenuates the development of morphine

tolerance and dependence suggests that chronic opioid treatment may elicit EAA

receptor dom-regulation or desensitization. Because opioid receptors and EAA

receptors have a sirnilar distribution in brain (Mansour et al, 1995; Masu et al, 1994; etc), and interact with the sarne intracellular second messenger systems, opioids may modulate EAA receptor activity, or vice versa, via actions on common pools of intracellular second messengers. m. Opioids and EAA-linked second messengers

A. Cyclic adenosine 3 :5'-monophosphate (CAMP) production

Since the 1970's. experimental observations suggest that opioids interact with the adenylate cyclase/cAMP intracellular second messenger system. There is a large selection of both behaviourai and biochernicd data.

Manipulating levels of CAMP has been shown to influence the behavioural and neuronal depressant effects of opioids. For example, opioid-induced antinociception has been shown to be inhibited by treatments which increase CAMP. Directly increasing CAMP levels by pre-treatment with CAMP or dibutyryl CAMP (BtZcAMP,a

CAMP analogue) attenuated the analgesic effect of p- and 6opioid receptor agonists

(Ho, Loh, and Way, 1972; Wang, Ren, Xue and Han, 1993), but did not affect K- receptor agonist-induced antinociception (Wang et al, 1993). Preventing the conversion of CAMP to adenosine monophosphate (AMP) by adrninistering p hosp hodiesterase inhibitors such as methylxanthines (34sobutyl- 1-methylxant hine

(IBMX), caffeine, theophylline) and Ro 20- 1724 (Ho et al, 1972; Nicholson, Reid and

Sawynok, 199 1), and stimulating CAMP production with forskolin @Iicholson et ni,

199 1) also inhibited the analgesic effects of opioid agonists. However, it should be noted that Nicholson et al (1991) noted a dual effect of phosphodiesterase inhibitors and forskolin, with low doses inhibiting and high doses enhancing morphine-induced antinociception. Increasing CAMP production with forskolin, or directly applying CAMP. BtzcAMP or dioctanoyl CAMP has been shown to prevent opioid depressant effects on dorsal hom neurons in fetal mouse spinai cord-ganglion explants (Crain,

Crain and Peterson, 1986).

Not only does manipulation of CAMP levels affect opioid activity, but administration of opioids also affects adenylate cyclase activity and CAMPproduction.

Acute administration of p- and 6-opioid receptor agonists has generally been shown to decrease adenylate cyclase adivity and CAMP production in vivo (Clouet, Gold and

Iwatsubo, 1975; Guitart, Thompson, Mirante, Greenberg and Nestler, 1992;

Przewlocka, Dziedzicka, Lason and Przewlocki, 1992; Wang et al, 1993), in NG 108-

15 ce11 cultures which express predominantly 6-opioid receptors (Law and Loh, 1986;

Sharma, Nirenberg and Klee, 1975), in SHSYSY ce11 cultures which express primarily p-opioid receptors (Carter and Medzihradsky, 1993; Lambert, Atcheson, Hirst and

Rowbotham, 1993; Yu, Eiger, Duan, Lameh and Sadee, 1990), in cloned rat p-opioid receptors expressed in COS and CHO cells (Johnson, Wang, Wang and ühl, 1994) and in spinal cord and brain tissue slices and ce11 cultures (Bhoola and Pay, 1986;

Crain et al, 1986; Eriksson, Hansson and Ro~back,1992; Hejna, Hogenboom,

Portoghese and Mulder, 1989; Schoffelmeer, Nerheijden, Hogenboom, Mulder and

Stoof, 1987; Schoffelmeer, Rice, Jacobsen, Van Gelderon, Hogenboom, Hejna and

Mulder, 1988). The relationship between K-opioid receptors and CAMP production is less clear. Some investigators have found that the interaction between K-opioid receptors and the adenylate cyclase/cAMP system is similar to the interaction seen with p- and 6-opioid receptors (Bhoola and Pay, 1986; Eriksson et al, 1992). Other investigators observed no interaction between K-opioid receptors and the adenylate cyclase/cAMP system (Przewlocka et 41, 1992; Wang et al, 1993; Hejna et al, 1989).

Changes in adenylate cyclase activity and CAMP production have also been implicated in the development of opioid tolerance and dependence. Mer chronic administration, the ability of opioid agonists to inhibit CAMP production has been shown to be decreased (Carter and Medzihradslq, 1993; Eriksson et al, 1992; Law and Loh, 1986; Van Vliet, Van Rijsqijk, Wardeh, Mulder and Schoffelmeer, 1993) and basal CAMPproduction may be increased (Tenviiiiger, Beitner-Johnson, Sevarino,

Crain and Nestler, 199 1). In addition, adenylate cyclase activity and CAMP production has been shown to increase towards pre-treatment levels with chronic administration of opioid agonists. During chronic morphine treatment, there are increased levels of adenylate cyclase mRNA in locus coemleus and amygdala

(Matsuoka, Maldonado, Defer, Noel, Hanoune and Roques, 1994). Furthemore, chronic morphine treatment increases the expression of CAMP response element

(CRE)-binding protein (CREB) immunoreactivity (CREB mRNA) in the locus coeruleus in vivo, and in a locus coeruleus-like ce11 line in vitro (Widnell, Russell and

Nestler, 1994). In another study, it was demonstrated that acute administration of morphine decreases the phosphorylation state of CREB in rat locus coemleus, while during chronic morphine treatment the effect on CREB was attenuated, and during withdrawal the phosphorylation of CREB was increased (Guitart et al, 1992).

Increased activity of adenylate cyclase in various brain regions and dorsal root gangliodspinal cord @RG-SC) CO-culturesfiom chronic morphine treated rats has also been indicated by increased levels of Gia, Goa, adenylate cyclase, CAMP-

dependent protein kinase and certain CAMP-related phosphoproteins (TeMlilliger et al,

199 1).

Treatments which affect the level of CAMPcan dso influence the development

of tolerance and dependence. Repeated administration of CAMPduring morphine

treatment increased the rate of development of tolerance to morphine's analgesic

effects (Ho et al, 1972). Treatment with a combination of the methylxanthine

phosphodiesterase inhibitor [BMX and heroin accelerated the development of

dependence to heroin (Collier, Cuthbert and Francis, 198 1). Pre-treatment with

CAMPjust pnor to the precipitation of withdrawal and repeated administration of

CAMP during chronic morphine treatment increased severity of withdrawal symptoms

(Ho et al, 1972). Pre-treatment with methylxanthines or other phosphodiesterase

inhibitors just pnor to the precipitation of withdrawal also increased the severity of

abstinence symptoms (Collier and Francis, 1975). Moreover, methylxanthines and

other phosphodiesterase inhibitors when given alone can produce a "quasi-withdrawal"

syndrome where rats or mice exhibit withdrawal-like behaviours (Collier et al, 198 1).

B. Ph osp hatidy linositof hydrolysis

As discussed earlier, activation of group 1 mG1uR.s stimulates PLC mediated PI

hydrolysis, leading to the production of DAG and IP3.DAG promotes the translocation and activation of PKC, and Pa stimulates the release of ca2' from intemal stores on the endoplasrnic reticulum. Recent experimental evidence suggests that there are interactive effects between opioids, glutamate and the products of PI hydrolysis, as well as opioid interactions with the adenylate cyclase/cAMP system. It was shown that both acute application of the p-opioid agonist ~-~la*-~e~hd-~l~- o15-enkephalin(DAMGO) and administration of PKC potentiated glutamate activated currents in cultures of spinal trigeminal neurons in thin medullary slices from rats in a similar manner (Chen and Huang, 199 1). The effeas of both PKC and DAMGO were attenuated by protein kinase C (PKC) inhibitors (Chen and Huang, 1991). suggesting that activation of p-opioid receptors may induce PKC activation.

Release of endogenous opioids can be influenced by activation of the PI system. Recently, it has been shown that PKC activators can stimulate secretion of j3- endorphin from cultured hypothalamic cells (Kapcala, Weng and Juang, 1992). In the anterior pituitary, the secretion of the precursor of p-endorphin, pro-opiomelanocortin

(POMC) can be stimulated by activation of PKC (Abou-Samra, Hanvood, Catt and

Aguilera, 1987).

In addition, opioid-induced analgesia cm be influenced by alterations in PI hydrolysis. The analgesic efficacy of severai p-opioid agonists (morphine, DAMGO and sufentanil) was reduced by pre-treatment with lithium chloride (Raffa, Connelly and Martinez, 1992; RafFa and Martinez, 1992). Lithium chlonde inhibits the activity of the enzyme inositol- 1-phosphatase, thereby blocking resynthesis of PIPz from inositol phosphate (IP) and decreasing PI hydrolysis (Li, Sibony, Green and Warsh,

1993; Manji, Etchebemgary, Chen and Olds, 1993). It was hrther found that treatment with IP, restored the efficacy of andgesics in rnice pre-treated with lithium chloride (RaEa and Martinez, 1992). These results suggest that opioid analgesia may be mediated at least in part by increased PI hydrolysis. However, other investigators

(Zhang, Wang and Han, 1990) observed that PKC activation attenuated pl6- and K- opioid agonist-induced anaigesia. Thus, the interaction between opioids and the PI system rnay be cornplex.

PI hydrolysis can also be measured following opioid administration. A decreuse in PI hydrolysis is generally observed upon acute administration of opioids when using models in which p-opioid receptors are prevalent, or when selective p- opioid agonist s are used (Barg, B elcheva, Zirnlichrnan, Levy, Saya, McHale, JO hnson, coscia and Vogel, 1994; Barg, Belcheva and Coscia, 1992; Johnson, Wang, Wang and

Uhl, 1994). However, other investigators have observed a stimulation of PI hydrolysis upon administration of low doses (nM) of p-opioid agonists (Smart and Lambert,

1996; Smart, Smith and Lambert, 1994; Srnart, Smith and Lambert, 1995; Jin, Loh and

Thayer, 1992). only observing inhibition of PI hydrolysis upon administration of high

(ph&) concentrations (Srnari and Lambert. 1996). In contrast, in systems in which 6- or K-opioid receptors are predominant, or when selective 6- or eopioid agonists are used, enhanced PI hydrolysis is usually observed (Barg,Nah, Levy, Saya and Vogel,

1993 ; Feng, Nariata, Makimura, Hoskins and Ho, 1994; Jin, Lee, Loh and Thayer,

1994; Leach, Shears, Kirk and Titheradge, 1986; Okajima, Tomura and Kondo, 1993;

Periyasamy and Hoss, IWO; Smart et al, 1994; Tsu, Chan and Wong, 1995; but see also Yu and Sadee, 1986).

Recently, the effects of chronic opioid treatment on PI hydrolysis have been examined. PI hydrolysis has been measured dunng both the toleraddependent state (prior to the precipitation of withdrawal) and during the withdrawai state. Some investigators have found that in a tolerantldependent state rat (Busquets, Excnba,

Sastre and Garcia-Sevilla, 1995; Dkon,Ting and Chang 1990) and human (Busquets el al, 1995) brains have decreased PI hydrolysis compared to normal brains. Other investigators (Pellegrini-Giampietro, Ruggiero, Giannelli, Chiarugi and Moroni, 1988) obsemed that PI hydrolysis in cortical slices taken from morphine-dependent rats was not different from PI hydrolysis in cortical slices taken from non-dependent rats. Yet others (Nanta, Makimura, Feng, Hoskins and Ho, 1994b) found increased PI hydrolysis and PKC activity in rats treated chronically with morphine. In these studies.

PI hydrolysis after chronic opiate treatment was compared only to untreated controls.

If acute opioid treatment produces a higlûy significant decrease in PI hydrolysis, then in chronic opioid-treated rats PI levels which are oniy slightly lower or slightly greater than control values may in fact represent a compensatory increase in activity of the system. Under wit hdrawai conditions, there is general agreement t hat PI hydrol ysis is greatly enhanced (Busquets et al, 1995; Pellegrini-Giampietro el al, 1988). These results suggest that there are significant eEects of chronic opioid administration on PI hydrolysis. It is proposed that chronic morphine treatment elicits compensatory mechanisms in the central nervous system, with a resultant over-compensation in PI hydrolysis during the withdrawal state.

Recent evidence suggests that chronic opioid treatment may affect PI hydrolysis in the spinal cord as well as the brain. Mao, Price, Phillips, Lu and Mayer

(1995) observed that although acute intrathecal (i.t.) administration of morphine did not affect PKC activity in the spinal cord dorsal hom of rats, chronic i.t. morphine

increased PKC activity .

it has also recently been show that the development of opioid tolerance and

dependence can be attenuated with protein kinase inhibitors. 1.c.v. infusion of H-7, a

non-selective protein kinase inhibitor (Hidaka, Inagaki, Kawamoto and Sasaki, 1984),

concurrently with i.c.v. morphine or butorphanol, attenuated the development of

analgesic tolerance to i.c.v. administered morphine or butorphanol (Narita, Feng,

Makimura, Hoskins and Ho, 1994a). Furthemore, it has also been shown that

coadministration of GM 1 ganglioside with intrathecal morphine attenuates the

development of tolerance and dependence (Mao, Price and Mayer, 1994; Mayer, Mao

and Price, 1995). However, as with H-7, although GMI ganglioside does inhibit the

translocation of PKC, its effects are not selective. GMI ganglioside also inhibits the

activation of phospholipase A2 (Hungund, Zheng, Lin and Barkai, 1994), and

modulates the activity of calcium charnels (Carlson, Masco, Brooker and Spiegel,

1994; Bressler, Belloni-Olivi and Forman, 1994). Thus, because H-7 and GM 1 ganglioside are not selective, it is not entirety clear that their ability to decrease the development of tolerance and dependence is due to effects on PKC.

The remaining chapters of this thesis further explore the relationship of EAA receptors, particularly rnGluRs, and related intracellular second messenger systems to the development of morphine tolerance and dependence. In Chapter 2, we present initial evidence for the involvernent of mGluRs, while in Chapter 3, we examine the role of specific mGluR subtypes in the development of morphine dependence. In Chapter 4, we fùrther explore the role of PI hydrolysis. In Chapter 5 we test the hypothesis that mGiuR desensitization may contribute to morphine dependence and withdrawal. In Chapter 6 we assess the role of gopioid receptors (which are positively coupled to PI hydrolysis) in the developrnent of both morphine tolerance and dependence. In the general discussion, our mode1 of the development of morphine dependence is cornpared to severd previously proposed models. CHAPTER 2: A closer look at the effects of excitatory amino acid antagonists

on morphine tolerance and dependence As discussed earlier, it has been show that chronic antagonism of NMDA receptors concurrently with morphine treatment attenuates the development of tolerance and dependence. The endogenous arnino acid glutamate acts at NMDA receptors. therefore glutamate may be involved in the development of opioid tolerance and dependence. In addition to NMDA receptors, glutamate acts at AMPA/kainate and metabotropic glutamate receptors (Mayer and Westbrook, 1987; Monaghan,

Bridges and Cotman, 1989). Thus, it is interesting to examine the contribution of t hese other glutamate receptors to the development of morp hine tolerance and dependence. We are the first to examine the behaviourd effects of AMPAkainate and mGluR antagonism on the developrnent of morphine tolerance and dependence.

We are pa.rticularly interested in the contribution of mGluRs for several reasons. First, the involvement of NMDA receptors in the developrnent of morphine tolerance and dependence has been well documented by several groups of investigators (Ben-Eliyahu, Marek, Vaccarino, Mogil, Sternberg and Liebeskind,

1992; Elliot, Mnarni, Kolesnikov, Pasternak and Intumsi, 1994; Koyuncuoglu et al,

1990; Marek rial, 19914b; Tiseo and Intumsi, 1993; Trujillo and Akil, 199 1; Trujillo and Akil, 1994), and activity at mGluRs modulates activity at NMDA receptors.

Several investigators have observed that application of mGluR agonists or PKC potentiates NMDA-mediated currents (Bleakman, Rusin, Chard, Glaum and Miller,

1992; Aniksztejn, Otani and Ben-Ari, 1992; Harvey and Collingridge, 1993; Kelso,

Nelson and Leonard, 1992; Chen and Huang, 1992). Thus, antagonism of mGluRs during chronic morphine administration rnay attenuate NMDA-receptor activity, thereby eliciting similar reductions in the developrnent of morphine tolerance and dependence as does antagonism of NMDA receptors. However, there is some controversy, as other investigators have observed an mGluR-mediated depression of

NMDA-induced currents and increases in intracellular [ca2+]via a PKC mechanism

(Colwell and Levine, 1994; Courtney and Nicholls, 1992).

Second, mGluRs are directly coupled, through G-proteins, to CAMP production and PI hydrolysis (Nakanishi 1992; Schoepp and Conn, 1993), as are opioid receptors (Chiiders, 199 1; Barg et al, 1992, 1993, 1994). Thus, we hypothesize that mGluR activity may modulate opioid activity by effects on these intracellutar second messenger systems and vice versa.

Third, brain distribution of mGluRs is similar to that of opioid receptors.

There is high expression of mRNA for both opioid receptors and rnGIuRs in thalamus, stnatum and conex (Mansour et al, 1995; Masu et al, 1994). Moreover, there is high expression of p-opioid receptors in the locus coeruleus (Mansour et al, 1995), and although mGluRs are not prominently expressed here. there are EAA projections from the nucleus paragigantocellularis (Akaoka and Aston-Jones, 199 1; Ennis, Aston-Jones and Shiekhattar, 1992). Because the distributions are similar, it can be hypothesized that mGluRs and opioid receptors may be CO-localizedwithin the same cells, lending credence to the hypothesis that mGluR activity may modulate opioid activity via effects on CAMP production andor PI hydrolysis.

Therefore, the first manuscript deals with the eEects of chronic antagonism of

NMDA, AMPAkainate and mGluR receptors on the development of morphine dependence. Because the majority of work examining the role of NMDA receptors in the development of opioid tolerance and dependence has been done using MK-80 1, we also used it as an NMDA antagonist to provide a direct cornparison with previous data. We used the non-competitive AMPA/kainate receptor antagonist 1-(4- arninophenyl)-4-methy1-7,8-methylenedioxy-SH-2,3-benzodiazepine-hydrochloride

(GYKT 52466) (Donevan md Rogawski, 1993), as opposed to the more cornmonly used 6-cyano-7-nitroquinoxaline-2,3 -dione (CNQX) or 6,7-dinitroquinoxdine-2,3- dione (DNQX) (Honoré, Davies, Drejer, Fletcher, Jacobsen, Lodge and Nielsen, 1988;

King, Lopez-Garcia and Cumberbatch, 1992), because GYKI 52466 has been shown to be stable in osmotic pumps for at least two weeks (Steppuhn and Turski, 1993), and because we observed toxic side effects of CNQX in pilot studies. To antagonize rnGluRs we first used L-2-amino-3-phosphonopropropanoic acid (Birse, Eaton, Jane,

Jones, Porter, Pook, Sunter, Udvarhelyi, Wharton, Roberts, Salt and Watkins, 1993;

Schoepp et al, 1990). However, since L-AP3 shows some activity at NMDA receptors (Birse et al, 1993), we also included a more selective mGluR antagonist, the phenylglycine derivative S-4-carboxyphenylglycine ((S)-4CPG), which is selective for group I mGluRs which are positively coupied to PI hydrolysis (Birse et al, 1993;

Eaton, Jane, Jones, Porter, Pook, Sunter and Udvarhelyi, 1993; Hayashi et al, 1994;

Watkins and Collingridge, 1994).

To examine the role of the different groups of EAA receptors in the development of morphine tolerance and dependence, we chronically infused EAA antagonists intracerebroventricularly (i. C.V.) concurrently wit h chronic systemic morphine infusion. We assessed the analgesia induced by the ifised morphine 4 to 6 hours following implantation of osmotic pumps containing morphine sulfate solution

(acute) by measuring response latency in the tail flick test. To determine whether tolerance had developed to the infused morphine, we again measured tail flick latency on the sixth day of treatment. To assess the degree of dependence, we measured the seventy of precipitated withdrawal symptoms on the seventh day of morphine treatment (see manuscript).

As discussed in the manuscript, chronic antagonisrn of either NMDA or mGluR receptors concurrently with morphine treatment decreased the severity of the precipitated withdrawal syndrome. Although there was a trend for EAA antagonist- treated rats to exhibit greater morphine analgesia on the sixth day of treatment, we did not observe a statistically significant effect on the development of tolerance (data not shown). Because we did not observe a significant effect on the development of tolerance, we concentrated our efforts on examining the development of dependence in future experiments involving EAA receptors. Moreover, we also did not observe an effect of acute i.c.v. administration of antagonists just prior to the precipitation of withdrawal (data not shown). Br. J. Pharmacol (199-1) 113: 1215-1220.

Morphine dependence is affected by activity at metabotropic,

as well as ionotropic (NMDA), glutamate receptors

1.2 1.23.4 Marian E. Fundytus and Terence J. Coderre

1 Pain Mechanisms Laboratory, Clinical Research Institute of Montreal , 2 Department of Psychology, McGill University , and Centre de recherche en sciences neurologiques et 3 Département de médecine, Université de Montréal .

Short rifle: EAA antagonists attenuate morphine withdrawal

4 To whom reprint requests should be addressed:

Terence J. Coderre, Ph.D. Pain Mechanisms Laboratory Clinical Research Institute of Montreal 1 10 Pine Ave. West, Montreai, (Quebec) Canada, WW 1R7 Summary

1. The contribution of various excitatory amino acid (EAA) receptors

OA,AMPAkainate and metabotropic) in the brain to the development of morphine dependence was exarnined. This was performed by measuring the seventy of the precipitated withdrawal syndrome following chronic subcutaneous (SC) morphine and intracerebroventncuiar (icv) EAA antagonist treatment.

2. Continuous subcutaneous (SC)treatment with morphine sulfate (36.65 pmol

- 1 day ) produced an intense and reliable naloxone-precipitated withdrawal syndrome.

3. Chronic icv treatment with antagonists selective to metabotropic and

NMDA receptors, but not AMPAntainate receptors, significantly attenuated abstinence symptorns. Conversely, EAA antagonists had very little effect on non- withdrawal behaviours.

4. These results suggest that as well as changes elicited by activation of

NMDA receptors, metabotropic receptors and intracellular changes in the phosphatidylinositol (PI) second messenger system or the cyclic adenosine 3',5'- monophosphate (CAMP) second messenger system, to which EAA metabotropic receptors are linked, may be involved in the development of opioid dependence with chronic morphine treatment.

Key words: opioid; morphine; de pendence; glutamate; metabotropic glutamate receptor; AMPA; kainate; NMDA, abstinence syndrome; excitatory amino acids Introduction

Although opioid drugs such as morphine are widely used for the management of pain, their clinical usefùlness is limited by the development of tolerance and dependence with their chronic use. Tolerance is indicated by a decreased efficacy of the drug with repeated administration, and results in a need to increase the morphine dose in order to achieve the desired analgesic effect. Dependence is a continued need for the drug to maintain a state of physiological equilibrium, and leads to an aversive withdrawal or abstinence syndrome when morphine administration is terminated.

Recently, it has been demonstrated that coadministration of N-methyl-D-aspartate

(NMDA) receptor antagonists attenuates the development of morphine tolerance and dependence (Marek et a!., 199 1a; Marek et al., 199 1b; Trujillo & Akil, 199 1). Since the endogenous excitatory amino acid (EAA) glutamate activates NMDA receptors, it is likely that glutamate contnbutes to the development of these phenornena. In addition to NMDA receptors, glutamate acts at at least two other types of ionotropic receptors: receptors at which a-2-arnino-3-(hydroxy-5-methylisoxazol-4yoic acid (AMPA) is a selective agonist and receptors at which kainate is a selective agonist; as well as at metabotropic receptors (Mayer & Westbrook, 1987a; Monaghan et al., 1989). While a role for the NMDA receptor has already been suggested, the specific contribution of these other receptors to behavioural indices of opioid tolerance and dependence has not been investigated.

21 The NMDA receptor gates a cation channel that is permeable to Ca and ~a*

(MacDermott et al., 1986; Mayer et al., 1987) and is gated in a voltage-dependent fashion by M;~ (Mayer et al., 1984; Nowak et al., 1984). It is the voltage-dependent

2- Ca pemeability of the NMDA receptor that is thought to be necessary for use- dependent synaptic plasticity (Cotman et al., 1988), and may be critical for the development of neuronal changes that mediate opioid tolerance and dependence

(Marek et al., 199 la, b; Trujillo & Akil, 199 1). AMPA and kainate receptors gate cation channels that are permeable to Na . but for the most part have negligible

3- permeability to ~a (Mayer & Westbrook, 1987b; Murphy & Miller, 1989). However.

AMPNkainate receptors have recently been cloned that exhibit C: ' permeability

(Miller, 1 99 1 ; Sommer and Seeburg, 1992). Traditiondly, AMPAkainate receptors are thought to be involved in the mediation of rapid excitatory responses to EAA transmitters (Jonas and Sakmann, 1992; Kiskin et al., 1986), and may contnbute to neuronal plasticity by relieving the NMDA receptor of its voltage-dependent block by

2- Mg . Unlike ionotropic receptors, metabotropic glutamate receptors are not linked to cation charnels. Instead they are coupled directly to the ce11 membrane by a G protein

(Sladeczek et al., 1985; Sugiyama et al., 1987). Several subtypes of metabotropic glutamate receptors have recently been cloned. Some subtypes affect phosphatidylinositol (PI) hydrolysis (mGluR 1a, mGluR 1P and mGluRS), while others affect the production of adenosine 3',5'-cyclic monophosphate (CAMP) (mGluR2, rnGluR3, mGluR4) (Schoepp & Conn, 1993). Activity at metabotropic receptors coupled to the PI system activates phospholipase C, which catalyzes phosphatidylinositol hydrolysis, leading to the production of inositol- 1,4,5- tnsphosphate (IP3) and diacylglycerol @AG) (Arnbrosini & Meldolesi, 1989; Manzoni el al., 1990; Schoepp & Conn, 1993). Activity at metabotropic receptors coupled to the CAMP second messenger system generaliy Ieads to decreased production of CAMP, dthough activation of mGIuRla stimulates an increase of CAMP

(Schoepp & Conn, 1993). Through the increased production of intracellular messengers associated with PI hydrolysis, or decreased production of CAMP. metabotropic receptor activation may play an important role in the long-terrn effects mediated by glutamate (Nicoletti et al., 199 l), and like NMDA receptors rnay be critical to the development of neuronal changes mediating opioid tolerance and dependence.

In the present sîudy we have investigated the contribution of various EAA receptor subtypes in the brain to the development of opioid dependence. This purpose was achieved by exarnining the effects of the intracerebroventricular (icv) administration of seiective EAA receptor antagonists concurrently with the chronic subcutaneous (SC) administration of morphine. NMDA receptors were antagonized with the noncornpetitive antagonist 5-methyl- l0,ll -dihydro-5H-dibenzocyclohepten-

5.1 O-irnine hydrogen rnaleate (MK-80 1) (Wong et al., 1986); AMPAfkainate receptors were antagonized with 1-(4-Aminopheny1)-4-rnethyl-7.8-methylenedioxy-5 H-2,3 - benzodiazepine-hydrochloride (GYKI 52466) (Donevan & Rogawski, 1993).

Although distinct families of high-affinity AMPA and kainate receptors have been isolated, the functional distinction between these receptors is not entirely clear

(Barnard & Hedey, 1990). Consequently, phannacological investigations with receptor antagonists are lirnited to investigations of non-selective AMPA/kainate receptor effects. Metabotropic receptors were antagonized with the highly selective antagonist (S)-4-carboxyphenylglycine ((S)-4C-PG) (Birse et ai., 1993; Eaton et ai.,

1993) and a more cornmoniy used, yet less selective antagonist, L-2-arnino-3- phosphonopropanoic acid (L-AP3) (Birse et al., 1993; Schoepp et al., 1990).

Methods

Subjecis and Swgery

Subjects were male Long Evans rats (280-350 grarns). The rats were housed 2 to 3 per cage, on a 12: 12 hour 1ight:dark cycle (lights on at 06:00), with food and water available ad libitum.

On Day O rats were anaesthetized with sodium pentobarbital (Sornnotol, MTC

Pharmaceuticals, 60 mg kg -'), and 23 gauge stainless steel cannulae, attached to

Model 2001 Alzet@ osmotic rnini pumps filled with one of the EAA antagonist solutions or saline, were implanted stereotaxically in the lateral ventncal (AP= -1.3 mm and L= - 1.8 mm From bregma and V= -3.8 mm fiom the top of the skull; Paxinos

& Watson, 1986). While the rats were still under pentobarbital anaesthesia, one

- I unprimed (Le. not yet pumping) Model 2MLl Alzeta pump containing 60 mg ml morphine sulfate solution was implanted SC on the back. These morphine containing pumps started pumping the morphine solution approximately 2 to 4 hours following implantation. On the following day, Day 1, rats were bnefly anaesthetized with halothane and a second unprimed Model 2MLl Ahet@ purnp containing 60 mg ml -' morphine sulfate solution was implanted SC on the back. This two day purnp implantation procedure was used to reduce the risk of mortality due to the accumulation of lethal systemic morphine concentrations pnor to any tolerance development. To assess the effects of chronic icv EAA antagonist treatment on behaviour in rats not dependent on morphine, some rats were given vehicle or 40 nmol

- 1 day of either L-AP3, (S)-4C-PG, MK-801 or GYKi 52466 without concurrent morphine treatment.

Dmgs

MK-80 1, L-AP3 and GYKI 52466 were obtained from Research Biochemicals,

Inc. (Natick, MA), while (S)-4C-PG was purchased from Tocris Neurarnin (Bristol.

-1 UK). EAA antagonists were continuously infùsed at a rate of 1 pl hr in the

- I -1 - 1 following concentrations: 1.6 nmol day , 8 nmol day and 40 nmol day . Morphine sulfate (Sabex, Montreal, PQ) was continuously delivered at a rate of 10 pl hr -' for a

-1 total dose of 36.65 pmol day morphine sulfate.

Withdrawai Measurement

Precipitated abstinence symptoms were assessed on the seventh day of treatment (while al1 pumps were still delivering antagonists and morphine) fier

- 1 injection of naloxone (1 mg kg , SC). For 10 minutes before and 40 minutes after naloxone injection, the withdrawal symptoms were assessed by measuring the amount of time spent teeth chattering and writhing, as well as by counting jumps and wet dog shakes. The time spent in non-withdrawal behaviours (arnbulating, rearing, grooming and resting) was also measured for cornparison, for 10 minutes before and after the injection of naloxone, in rats treated with icv EAA antagonists either alone or with SC morphine. Siatistzcal Analyss

Timed withdrawal behaviours (teeth chattering, writhing) were analyzed using

ANOVA, followed by post-hoc tests on significant main effects. Counted withdrawal behaviours (number of jumps and wet dog shakes) were analyzed using a Kmskal-

Wallis ANOVA for non-pararnetric data, followed by Mann-Whitney U-tests on significant main effects.

The effect of EAA antagonist treatment on non-withdrawal behaviours

(ambulating, rearing, groorning and resting) was assessed by cornparhg the first two time blocks (i.e. 10 minutes pnor to naloxone injection and 10 minutes fier naloxone injection) for rats in each treatment group. Planned cornparisons were used to analyze differences in the percent of tirne spent in each timed non-withdrawal and the two timed withdrawal behaviours dunng these two time blocks across the different treatment conditions.

Results

- 1 Administration of 36.65 pmol day SC morphine sulfate by AlzetB purnp produced an intense and reliable naloxone-precipitated abstinence syndrome which was evidenced by the occurrence of teeth chattering, writhing, jumping, and wet dog shaking. As indicated in Fig. 14 the metabotropic receptor antagonists (S)-4C-PG and L-AP3 significantly decreased the occurrence of timed abstinence symptoms (teeth chattering and writhing). The effect for (S)-4C-PG appeared to be dose-dependent, with the highest dose, 40 nmol day" producing the greatest reduction in teeth

- 1 chattering and writhing. L-AP3 was most effective at 8 nmol day . Fig. IB shows the arnount of time spent teeth chattering and writhing for rats treated with MK-801 and GYKl 52466. The NMDA receptor antagonist MK-801 significantly decreased the tirne spent in withdrawal at al1 doses used. The AMPNkainate receptor antagonist

GMU 52466 did not affect the amount of time spent in withdrawai at any of the doses used.

Figure 2 illustrates the fiequency of the counted abstinence symptoms, jumps and wet dog shakes. Although MK-801 tended to increase the number of jumps and wet dog shakes at the hi& dose (40 nmol day -'), none of the icv EAA antagonist treatments signrfiantiy affected the number of jumps and wet dog shakes.

Figure 3 shows the percent of time spent in each of the timed behaviours during the 10 minutes prior to naioxone administration, and dunng the 10 minutes following naloxone for rats in each icv treatment group either with or without concurrent morphine treatment. As can be seen in Figure 3A and 38, prior to the injection of naloxone, rats in al1 icv treatment groups, with or without morphine, behaved very similarly, with the only dflerences being more grooming in L-AP3 treated rats as compared to saline treated rats. In addition, saline treated rats that were also given morphine reared more than rats given icv saline alone. Although activity levels (ambulating and rearing) were lower and resting was generally higher after the injection of naloxone, rats given icv EAA antagonists without morphine still behaved very similarly to rats given icv saline without morphine. The oniy differences observed were an increase in grooming in L-AP3 treated rats and increased activity in

GW52466 treated rats as evidenced by increased ambulating, rearing and grooming, and decreased resting. As expected, rats dependent on morphine showed significantly more naloxone-precipitated withdrawal, with a resultant decrease in non-withdrawal behaviours compared to rats given icv treatments done. In morp hine-dependent rats.. time in withdrawal was significantly less in L-AP3, (S)-4C-PG and MK-80 1 treated rats, which coincided with an increase in arnbulation in (S)-4C-PG and MK-801 treated rats.

Discussion

The present results demonstrate that concurrent treatment of rats with various

EAA antagonists and chronic morphine leads to a decrease in various symptoms of morphine withdrawal. The NMDA receptor antagonist MK-80 1 and the metabotropic receptor antagonists (S)-4C-PG and L-AP3 were al1 effective at decreasing the amount of time spent exhibiting the withdrawal symptoms of teeth chattering and writhing, while the AMPNkainate receptor antagonist GYKI 52466 had no effect. Although al1 but the AMPAkainate receptor antagonist significantly reduced the timed withdrawal symptoms, none of the EAA antagonists significantly afEected the counted withdrawal symptorns (i.e.jumping and wet dog shakes). It has been suggested that withdrawal symptoms such as jumping are mediated primarily by structures around the fourth ventricle, as evidenced by focal brain micro-injections of naltrexone and levallorphan in morphine-dependent rats (Koob et al., 1992; Laschka et al., 1976). In the present study, EAA antagonists were infused in very small volumes into the lateral ventricle, thus it is possible that the counted symptoms were not affected because the drugs were unable to diffise to the appropriate brain structures around the fourth ventricle. There was very little effect of icv EAA antagonist treatment on non- withdrawai behaviours, except for a generai increase in groorning in rats given L-AP3 and increased activity in GYKI 52466 treated rats given naloxone. In general, rats given icv treatment alone rested more during the latter 10 minutes (Le. after naloxone), with concurrent decreases in ambulation and rearing. This is likely not an effect of naloxone, but rather due to the fact that they had adequately explored the test box and were cornfortable in the environment (a phenornenon comrnon in untreated rats). Rats treated chronically with morphine exhibited withdrawal following the injection of naloxone, which caused a subsequent decrease in other tirned behaviours. L-AP3. (S)-

4C-PG and MK-801 al1 decreased withdrawal compared to saline in morphine treated rats. with a resultant increase in ambulation in rats given MK-801 + morphine and (S)-

4C-PG + morphine. The decrease in withdrawal allows for more time to be spent in other behaviours.

Although we have no explanation for why L-AP3 would increase grooming, it appears to be a robust effect since it occurred in every condition except in rats going through withdrawal (i.e. morphine + naloxone). The increase in activity produced by

GW52466 appears to be less robust since it occurred only in the group also treated with naloxone. Nonetheless, it is not expected that these behavioural effects of L-AP3 or GYEU 52466 alter our conclusions about the effects of EAA antagonists on withdrawal behaviour. GYKI 52466 did not significantly affect withdrawal, and the effects of the metabotropic receptor antagonist L-AP3 were confirmed by another metabotropic antagonist, (S)-4C-PG, which did not significantly influence non- withdrawal behaviours. Therefore, we propose that EAA antagonist treatment affects the development of dependence directly, and not indirectly by interfenng with the measurement of the withdrawal behaviours.

Previously, it has been observed that concurrent treatment of rats with the non- selective EAA antagonist kynurenic acid (Marek et al, 199 1 a) or the non-cornpetitive

NMDA antagonist ML80 1 (Marek et al., 199 1b; Trujillo & Akil, 199 1) with daily injections of morphine attenuated the development of tolerance to morphine's analgesic effects. MK-80 1 also alleviated the severity of some symptoms of the precipitated withdrawal syndrome (Trujillo & Akil, 1991). Furthemore. some investigators have found that acute treatment with kynurenic acid and MK-80 1 only on the day of testing (i.e. not concurrently with morphine) to be effective in decreasing the severity of some withdrawal syrnptoms (Rasmussen et al., 199 la; Rasmussen et al., 199 1b; Tanganelli et al., 199 1), while others did not find this acute administration effective (Trujillo & Akii, 1991). While each of these above studies assessed the effects of sjwfemically adrninistered EAA antagonists, in additional experiments

(unpublished data), we have found that acute icv injections of EAA antagonists on day

7 pnor to precipitation of withdrawal failed to attenuate severity of abstinence symptoms, lending support to the latter group.

Hyperactivity in the locus coeruleus (LC) has been shown to be correlated with the morphine withdrawal syndrome (Aghajanian, 1978; Valentino & Wehby, 1989).

Central administration of non-specific, NMDA selective and AMPA/kainate selective

EAA antagonists into either the lateral ventricle or the locus coeruleus was found to decrease the hyperactivity of locus coeruleus neurons during precipitated morphine withdrawal (Akaoka & Aston-Jones, 1991), with the best effects produced by the non- specific EAA antagonist. Conversely, systemic administration of selective NMDA receptor antagonists was unable to affect the hyperactivity of locus coemleus neurons during precipitated morphine withdrawal (Rasmussen et al., 199 1a).

The present data indicates that chronic icv administration of the selective metabotropic EAA receptor antagonists (S)-4C-PG and L-AP3 was at least as effective in attenuating the severity of the morphine withdrawal syndrome as antagonists selective to the NMDA receptors. It is noteworthy that the most effective

-1 treatment was 8 nmol day L-AP3. This strong effect of L-AP3 rnay have resulted fiom a possible additive effect of L-AP3 at both metabotropic and NMDA receptors.

There is evidence that L-AP3 may have non-selective effects at the NMDA receptor as well as its proposed major action at the metabotropic receptor (Birse et al., 1993).

The significant dose-dependent effects of the highly selective metabotropic receptor antagonist (S)-4C-PG, however, do suggest an important role of metabotropic glutamate receptors in the development of morphine dependence. The failure of

GYKI 52466 to influence precipitated withdrawal suggests that AMPAkainate receptors do not play an important role in morphine dependence. Although it is possible that GYKI 52466's ineffectiveness could be explained by rapid metabolism IH vivo, this is unlikely because the dmg was chronically infûsed. Furthemore, it has been demonstrated, using an osmotic pump infusion method sirnilar to that used in the present study, that GYKI 52466 was as effective at attenuating excitatory amino acid induced seizures on the fourteenth day of infusion as it was on the third day of infusion

(Steppuhn and Turski, 1993).

Activity at specific metabotropic receptor subtypes stimulates p hosp hatidylinositol (PI) hydrolysis and leads to the production of the intracellular messengers inositol- 1,4,5- trisphosphate (Pj)and diacylglycerol @AG) (Ambrosini &

Meldolesi, 1989; Manzoni et al., 1990; Schoepp & Conn, 1993). Chronic opioid use may alter production of these intracellular messengers and thus elicit long-term changes which contribute to opioid tolerance and dependence. There is evidence that acute morphine treatment stimulates PI hydrolysis (RafFa & Martinez, 1992), while chronic morphine treatment inhibits PI hydrolysis (Dixon et al., 1990). (S)-4C-PG and

L-AP3 antagonism of the metabotropic glutamate receptor during morphine treatment may prevent cellular changes associated with persistent phosphatidylinositol hydrolysis, and consequently reduce withdrawal symptoms that are associated with these cellular changes. Other metabotropic receptor subtypes inhibit production of

CAMP (Schoepp & Corn, 1993). It is well established that both acute and chronic opioid use also affect the production of CAMP (Collier, 1980; Collier, 1983; Sharma el al., 1975). Thus, it is possible that (S)4C-PG and L-AP3 antagonism of metabotropic receptors prevented changes in the CAMP system associated with chronic opioid use.

Currently, we are detemùning whether changes in the PI system or changes in the

CAMP system associated with activation of metabotropic receptors are imponant for the development of tolerance and dependence with chronic opioid use. Thus, the present study indicates that both NMDA and metabotropic glutamate receptors rnay be involved in the development of dependence with chronic morphine use. Both NMDA and metabotropic glutamate receptors are associated with changes in intra-cellular second messenger systems. It is therefore hypothesized that NMDA and metabotropic glutamate receptor antagonists are effective because they prevent changes in second messenger systems associated with chronic opioid use.

Acknowledgements

This work was supported by gants fiom The Medical Research Council of

Canada (MT-1 1045) and Fonds de la Recherche en Santé du Québec (900051) to

T.J.C. References

AGFLUANIAN, G.K. (1978). Tolerance of locus coeruleus neurones to morphine and

suppression of withdrawal response by clonidine. Nature. 267, 186- 189.

AKAOKA. H.& ASTON-JONES,G. (199 1 ). Opiate withdrawal-induced

hyperactivity of locus coemleus neurons is substantially mediated by augrnented

excitatory amino acid input. J. Neurosci., 11, 3 830-3 839.

AMBROSINI,A. & MELWLESI,J. ( 1989). Muscarinic and quisqualate receptor-

induced phosphoinositide hydrolysis in primary cultures of striatal and hippocampal

neurons. Evidence for differential mechanisms of activation. J. Neurochem., 53,

825-833.

BARNARD,E.A. & HENLEY,J.M. (1990). The non-NMDA receptors: types, protein

structure and molecular biology. In The Pharmcology of ficitatory Amitzo

Acids, u TiPS SpeciuI Report, Eds. Lodge, D. and Collingridge, G. (Elsevier,

Amsterdam), 82-89.

BIRSE,E.F., EATON,S.A., JANE, D.E., JONES, P.L., PORTER,R.H., POOK, P.C.,

SUNTER, D.C.,UDVARHELYI, P.M., WHARTON,B., ROBERTS,P.J., SALT,T.E. &

WATKINS,J. C . ( 1993). Phenylglycine derivatives as new pharmacological tools

for investigating the role of metabotropic glutamate receptors in the central

nervous system. Neuroscience, 52, 48 1-488.

BLASIG,J., HERZ, A., RE~LD,K. & ZIEGLG~SBERGER,S. (1973). Development

of physical dependence on morphine in respect to tirne and dosage quantification

of the precipitated withdrawal syndrome in rats. Psychopharmacol., 33, 19-3 8. COLLER,H.O. J. (1 980). Cellular site of opiate dependence. Nature. 283, 625-629.

COLLIER,H.O.J.(1983). Physiologicd basis of opiate dependence. Drug & Aicohol

Dep., 111, 15-21.

COM, C.W., MONAGF~AN,D.T., & GANONG,A.H. ( 1988). Excitatory amino

acid neurotransrnission: NMDA receptors and Hebb-type synaptic plasticity.

Ann. Rev. Neurosci., 11, 61-80.

DIXON,W., TING,Y-W & CHANG,P.L. (1 990). The effects of morphine on

norepinephrine-stimulated phosphatidylinositol response in rat cerebral cortex.

Prog. Clin. Biol. Res., 328, 279-282.

DONEVAN,S.D. & ROGAWSKI,M.A. (1993). GYKI 52466, a 2,3-bentodiazepine, is

a highly selective, non-cornpetitive antagonist of AMPAkainate receptor

responses. Nerrrori, I Ou),5 1 - 59.

EATON,S.A., Jm,D.E., JONES, P.L., PORTER,R.H., POOK, P.C., SUNTER,D.C. &

UDVARHELY~,P.M. (1993). Cornpetitive antagonism at metabotropic glutamate

receptors by (S)-4-carboxyphenylglycine and (RS)-alpha-methyl-4-

carboxyphenylglycine. Eirr. J. Pharmacol., 244, 1 95- 197.

JONAS, P. & Sw,B. (1992). Glutamate receptor charnels in isolated patches

from CA 1 and CA3 pyramidal cells of rat hippocampal slices. J. PhysioL, 455,

143-171.

Kis~nu,N.I., KRISHTAL,O. A. & TSYNDF~JKO,A.Y. (1986). Excitatory amino acid

receptors in hippocampal neurons: Kainate fails to desensitize them. Nezrrosci.

Letr., 63, 225-230 . KOOB,G.F., MALDOMDO, R. & STNUS, L. (1992). Neural substrates of opiate

withdrawal. TrendrsNeurosci., 15, 186- 19 1.

LASCI-IKA, E., TESCKEMACHER,HJ., MEHRAEIN, P. & HERZ, A. (1976). Sites of

action of morphine involved in the development of physicai dependence in rats.

Psychophmucologia (Berl.), 46, 14 1- 147.

MACDERMO~,A.B., MAYER,M.L., WESTBROOK,G.L., SMITH,S.J. & BAREER,

I.L ( 1986). NMDA-receptor activation increases cytopiasmic calcium

concentration in cultured spinal cord neurones. Nature, 321, 5 19-522.

MANZONI,O. J. J., FMIELS-MARLIER,F., SAS SE^, I., BLOCKAERT,J., LE PEUCH,C.

& SLADECZEK,F.A.J. (1990). The glutamate receptor of the Qp-type activates

protein kinase C and is regulated by protein kinase C. Neurosci. Letf., 109, 146-

151.

WK,P., BEN-ELIYAHU, S., GOLD,M. & LIEBESKIM),J.C. (199 la).

Excitatory amino acid antagonists (kynurenic acid and MK-80 1) attenuate

the development of morphine tolerance in rats. Bruit1 Res., 547, 77-8 1.

MAREK, P., BEN-ELIYAHU,S., VACCARINO,A.L. & LIEBESIUND,J-C. (199 1b).

Delayed application of MK-80 1 attenuates development of morphine tolerance in

rats. Brain Rcs., 558, 163-165.

MAYER,M.L., MACDERMOTT,A.B., WESTBROOK,G.L., SMITH, S.J. & BARKER,

J.L (1987). Agonist- and voltage-gated calcium entry in cultured mouse

spinal cord neurons under voltage clamp measured using Arsenazo III. J.

Nerrrosci., 7, 3 23 0-3244. MAYER, M.L. & WESTBROOK,G.L. (1 987a). The physiology of excitatory amino

acids in the vertebrate centrai nervous system. Prog. NeurobioL. 28, 197-276.

MAYER, M. L. & WESTBROOK,G.L. (1 98%). Permeation and block of N-methyl-D-

aspartic acid receptor channels by divalent cations in mouse cultured centrai

neurones. J. Physiol., 394, 50 1-527.

MAYER,M.L., WESTBROOK,G.L. & GUTHRIE,P.B. (1984). Voltage-dependent block by M;+ of NMDA responses in spinal cord neurones. Nature. 309,

26 1-263.

MILLER.R. J. ( 199 1). The revenge of the kainate receptor. Trends Neurosci. 14,

477-479.

MONAGHAN,D.T., BRIDGES,R.J. & CO=, C.W. (1989). The excitatory amino

acid receptors: Their classes, pharmacology and distinct properties in the function

of the central nervous system. Ann. Rev. Pharmacol. Toxicol., 29, 365 - 402.

MURPHY.S.N. & MILLER,R. J. (1989). Regulation of caft influx into striatal

neurons by kainic acid. J. Phannacol. Erp. ïher.. 249, 1 84- 193.

NICOLE~,F., CONDORELLI,D.F., DELL'ALBANI, P., ARONICA, E., CATANIA,MY.,

SORTINO,M.A. & REYMANN,KG. (1991). Metabotropic glutamate receptors

and neuroplasticity. In ficitatory Amino Acids, Fidia Research Foundation

Symposizon Series, Vol. 9, R.P. Simon, Ed. (Thieme Medical Publishers, New

York). pp. 167-170. NOWAK,L., BREGESTOVSKI,P., ASCHER,P., HERBERT,A. & PROCHIANTZ,A.

( 1984). Magnesium gates glutamate-activated channels in mouse central neurones.

Nature, 30 7, 462-465.

PUNOS, G. & WATSON,C. (1986). The rat brain in stereotaxic coordinates.

Acadernic Press Inc., San Diego, California.

RAFFA. R.B. & MARTINEZ, R.P. (1992). Morphine antinociception is mediated

t hrough a LiCI-sensitive, 1.3- restorable pathway . EZK J Phmacol., 2 15, 3 57-

358.

USMUSSEN.K., FULLER, R.W., STOCKTON, M.E., PERRY,K.W., SW~ORD,R.M. &

ORNSTEIN,P.L. (1991). NMDA receptor antagonists suppress behaviors but not

norepinephrine turnover or locus coemleus unit activity induced by opiate

withdrawal. Ezrr. J. Phamcol., 197, 9-16.

RASMUSSEN, K., KRYSTAL,J.H. & AGMANIAN, G.K.(199 1). Excitatory amino

acids and morphine withdrawal: differential effects of centrai and peripheral

kynurenic acid administration. Psychophannacobgy, 105. 508-5 12.

SCHOEPP.D.D. & CON, P.J. (1993). Metabotropic glutamate receptors in brain

function and pathology. Trenk Pharmacol. Sci., 14, 13-25.

SCHOEPP,D.D., JOHNSON,B.G., SMITH,E.C.R. & MCQUAID,L.A. (1990). Stereo

selectivity and mode of inhibition of phosphoinositide-coupled excitatory amino

acid receptors by 2-amino-3-phosphonopropanoic acid. Mol. Pharmacol., 38,

222-228. SHARM~S.K., NIRENBERG,M. & KLEE,W.A. (1 975). Morphine receptors as

regulators of adenylate cyclase activity. Proc. Nati. Acad Sci. (USA), 72,

590-594.

SLADECZEK, F., PIN, J.P., RECASENS,M., BOCKAERT, J. & WEISS,S. (1 985).

Glutamate stimulates inositol phosphate formation in stnatal neurones. Nafure.

317, 717-719.

SOMMER,B. & SEEBURG,P.H. (1992). Glutamate receptor channels: novel

properties and new clones. Trends Phannacol. Sci., 13, 29 1-296.

STEPPUHN,KG. & TURSKI,L. (1993). Diazepam dependence prevented by

glutamate antagonists. Proc. Nati. Acad. Sci. (USA), 90, 6889-6893.

SUGIYAMA, H.,ITO, I. & HIRoNo, C.(1 987). A new type of glutamate receptor

linked to inositol phospholipid metabolism. Nature. 325, 53 1-533.

TANGANELLI,S., ANTONELLI,T., MORARI, M., BLANCHI, C. & BEAM,L. (199 1).

Glutamate antagonists prevent morphine withdrawal in mice and guinea pigs.

Nezrrosci. Lett., 122, 270-272.

TRUJILLO.K.A. & AKIL, H.(199 1). Inhibition of morphine tolerance and dependence

by the NMDA receptor antagonist MK-80 1. Science, 251, 85-87.

VALENTINO,R. J. & WEHBY,R.G. (1 989). Locus coemleus discharge characteristics

of morphine-dependent rats: Effects of naltrexone. brai^ Res., 488, 126- 134.

WONG, E.H.F., KEMP, J.A., PRESTLEY,T., KNIGHT, A.R., WOODRUFF,G.N. &

IVERSEN.L.L. (1986) The anticonvulsant MK-80 1 is a potent N-methyl-D-

aspartate antagonist. Proc. Nad Acad Sci. (USA). 83, 7104-7 108. Figure 1

Mean time spent exhibiting withdrawal (teeth chattering and writhing) during the 40 minute withdrawal period in rats given chronic SC morphine and icv treatment with the metabotropic receptor antagonists (S)-4C-PG (n = 5 to 10 per dose) and L-

AP3 (n = 4 to 10 per dose) (A), the NMDA receptor antagonist MK-80 1 (n = 4 to 10 per dose) (B), and the AMPAIkainate receptor antagonist GYKI 52466 (n = 4 to 10 per dose) (B). ANOVA indicated significant efiects of (S)-4C-PG (F(3 22) = 5.124, p

< 0.01). L-AP3 (F(3,23) = 12.107, p < 0.01), and MK-801 (F(3 21) = 6.222. p <

0.0 l), but not GYKI 52466 (F(3 20) = 0.07 1, p > 0.05). Significant differences fiom the control group are indicated by asterisks (* p < 0.05. LSD t-test). oc------O 1.6 8 40 Dose (nmoi/day)

A GYKI 52466

Dose (nmoi/day) Figure 2

Frequency of counted symptoms (jumps and wet dog shakes) during the 40 minute withdrawai period for rats given chronic SC morphine and icv treatment with the metabotropic receptor antagonists (S)-4C-PG (n = 5 to 10 per dose) or L-AP3 (n

= 4 to 10 per dose), the NMDA receptor antagonist M.-801 (n = 4 to 10 per dose). or the AMPAkainate receptor antagonist GYKI 52466 (n = 4 to 10 per dose).

Kniskal-Wallis ANOVA for non-parametric data revealed no significant effects of any of the EAA antagonists. (S)-4C-PG: H(3 22) = 0.803, p > 0.05; MK-80 1 : H(3 7 1) 7 7 -

= 0.747, p > 0.05; GYKI 52466: H(3 20) = 2.580, p > 0.05; L-AP3: H(3,23) = 7

2.142, p > 0.05.

Figure 3

Percent of time in non-withdrawai and withdrawal behaviours in rats treated

chronically with either saline, or 40 nmol day -l L-AP3, (S)-4C-PG,M.-801 or

GYKI 52466 (EAA antagonists) icv alone (i.e. no SC morphine; n = 6 to 8 per icv

treatment) (A) or with SC morphine (n = 4 to 10 per icv treatment) (B) during the 10

minutesprior to the injection of naloxone; and in rats given chronic icv saline or EAA

antagonists alone (C) or with SC morphine @) during the 10 minutes afrer the injection

of naioxone. Pnor to the injection of naloxone, rats in ail icv treatment groups, with

or without morphine, behaved very similady, with the only differences being increased

grooming in L-AP3 treated rats (planned comparison (1.52). p < 0.05) (A and B).

With the addition of morphine, saline treated rats exhibited more rearing than rats not

treated with morphine (planned comparison (1,52), p < 0.05) (B). Mer the injection

of naloxone. rats not dependent on morphine generally rested more than in the first 10

minutes. Non-dependent rats in al1 icv treatment groups behaved similarly, with the

only differences being more grooming in L-AP3 treated rats (planned comparison

(1.52). p < 0.05) (C), and more activity in GYKI 52466 treated rats as evidenced by

significantly more ambulating, rearing and grooming, and significantly less resting

(planned cornpanson (1 521, P c 0.05) (C). Morphine-dependent rats showed a

significant increase in withdrawal behaviours regardless of icv treatment after the

injection of naloxone (planned comparison (1 52). p < 0.05) @). However, L-AP3,

(S)-4C-PG and MK-801 al1 significantly decreased the percent of time spent in withdrawal behaviours, with a concurrent increase in ambulation in (S)-4C-PG and

MK-80 1 treated rats (planned cornparison (1 52). p < 0.05)@). a A No morphine: pre-naloxone B + morphine: pre-naloxone

SAL LAP3 (S)4CPG MEC-801 GYKI SAL LAP3 (S)4CPG MIC-801 GYKI

C No morphine: post-naloxone D + morphine: post-naloxone

SAL GAP3 SAL LAP3 (SmMK-801 GYRI CEIAPTER 3: A further examination of the contribution of mGluR subtypes to

the development of morphine dependence As discussed in the previous chapter, non-selective antagonism of mGluRs with L-AP3, as well as selective chronic antagonism of mGluRl and 5, which are positively coupled to PI hydrolysis, with (S)-4C-PG attenuated the developrnent of morphine dependence. In addition to non-selectively antagi>nizingal1 rnGIuRs, L-AP3 has been shown to have effects at NMDA receptors (Bine et al, 1993). Also, (S)-4C-

PG has a secondary action whereby it is an agonist at rnGluR2 and 3, which are negatively coupled to CAMP production. Therefore. it is not entirely certain From our first study whether inhibition of PI hydrolysis, or alterations in CAMP production was predorninantly responsible for the ability of either L-AP3 or (S)-4C-PG to decrease morphine dependence.

As discussed in the introduction, mGluRs can be divided into several classes based on receptor pharmacology and signal transduction mechanisms (Hayashi et ai,

1994). Group 1 mGluRs, which include mGluRl and 5, are positively coupled to PI hydrolysis (Schoepp and Conn, 1993). Group L1 rnGluRs (mGluR2 and 3), and group

III mGluRs (mGluR4, 6, 7 and 8) are negatively coupled to CAMP production

(Hayashi et al, 1994). Group II and DI mGIuRs are generally differentiated based on receptor phannacology (Hayashi et al, 1994; Saugstad et al, 1994; see Introduction).

Until recently, the most cornrnonly used mGluR antagonist available was L-2- amino-3-phosphoprionic acid (L-AP3). However, L-AP3 has some activity at NMDA receptors (Birse et al, 1993; Schoepp et al, 1990), and phenylglycine derivatives have become the antagonists of choice. The non subtype-selective phenylglycine antagonist

(+)-a-methyl-4-carboxyphenylglycine(MCPG) antagonizes group 1, II and III rnGluRs (Eaton, Jane, St-J. Jones, Porter, Pook, Sunter, Udvarhely, Roberts, Salt and Watkins.

1993; Hayashi et al, 1994; Watkins and Coliingridge, 1994; Thomsen, Boel and

Suzdak, 1994). MCPG has been shown to have no effect on ionotropic glutamate receptors at concentrations up to I mM in vitro (Thomsen et al, 1994), and is therefore highiy selective for mGluRs. Group II mGiuRs can be selectively antagonized with 2S, 1'S,2'S-2-rnethyl-2-(2'-carboxycyclopropyl)-ycine (MCCG)

(Jane, Jones, Pook, Tse and Watkins, 1994; Salt and Eaton, 1995). MCCG does not affect group 1 or III mGIuRs at concentrations up to 300 mM in vitro (Jane et al,

1994), and has no effect on either NMDA or AMPAhinate receptors at concentrations up to 500 mM (Jane et al, 1994). Group III rnGluRs can be selectively antagonized with a-methyl-L-amino-4-phosphonobutanoate (MAP4) (Jane et al,

1994; Salt and Eaton, 1999, which has no effect on other mGluRs at concentrations up to 300 mM N1 vitro (Jane et al, 1994), and no effect on ionotropic glutamate receptors at concentrations up to 500 rnM (Jane el al, 1994).

Because there are different classes of mGIuRs, and (S)-4C-PG has dual effects, and particularly since opioids interact with the adenylate cyclase/cAMP system, we were interested in fùrther charactering the contribution of CAMP linked mGluRs to the development of morphine dependence. Thus, we chronically infused either

MCPG, MCCG or MAP4 i.c.v. concurrently with systemic morphine infusion, or gave an acute i.c.v. injection of these agents 10 minutes prior to naloxone injection, and measured the severity of precipitated withdrawal symptoms on the seventh day of treatment. Decreased severity of abstinence in antagonist-treated rats would suggest that group II and III mGluRs are involved in the development of morphine dependence.

The next study shows that chronic antagonism of group II and III, as well as group 1, mGluRs, attenuates the development of morphine dependence. Acute i.c.v. administration of mGluR antagonists just prior to the precipitation of withdrawal failed to decrease the severity of abstinence symptoms, and in fact, MCCG significantly increased the time spent teeth chattering and writhing. Br. J. Phannacol. (submitled)

Atîenuation of morphine witbdrawal symptoms by subtype selective

metabotropic glutamate receptor antagonists

Marian E. ~und~tus",Jennifer ~itchie'& Terence J. ~oderre'.' '.

1Pain Mechanisrns Laboratory, Clinical Research hstitute of Montreal, 2~epartmentof Psychology, McGili University and 'centre de recherche en sciences neurologiques et département de médecine, Université de Montréal.

Short title: mGluR antagonists attenuate motp hine withdrawd

4To whom correspondence should be sent:

Dr. Terence J. Coderre Pain Mechanisms Laboratory Clinicai Research Institute of Montreal 1 10 Pine Avenue West Montreal, Quebec, Canada H2W I R7

Telephone: (5 14) 987-5750 FAX: (5 14) 987-5624 Summary

1. We have previously show that chronic antagonism of group 1

metabotropic glutamate receptors (rnGluRs), in the brain, attenuates the precipitated

morphine withdrawal syndrome in rats. In the present investigation we assessed the

effects of chronic antagonisrn of group II and III rnGluRs on the severity of

withdrawal symptoms in rats treated chronicaily with subcutaneous (s.c.) morphine.

2. Concurrently with S.C. morphine we infused intracerebroventricularly (i. c.v. )

one of a series of phenylglycine derivatives selective to specific mGluR subtypes.

Group II mGluRs (mGluRz3), which are negatively coupled to CAMP production,

were selectively antagonized with MCCG. Group III mGluR's (mGluIt. 6.7 and 1),

which are also negatively linked to CAMP production, were selectively antagonized

with MAP4. The eEects of MCCG and MAP4 were compared with MCPG, which

non-selectively antagonizes group II mGluRs, as well as group 1 mGluRs (rnGl~R,,~)

which are positively coupled to PI hydrolysis.

3. Chronic i.c.v. administration of both MCCG and MAP4 significantly

decreased the time spent in withdrawal, MCPG and MCCG reduced the frequency of jumps and wet dog shakes, and MCPG and MCCG attenuated the severity of

agitation.

4. Acute i.c.v. injection of mG1u.R antagonists just pnor to the precipitation of

withdrawal failed to decrease the severity of abstinence symptoms. Rather, acute i.c v.

injection of MCCG significantly increased the time spent in withdrawal. 5. Our results suggest that the development of opioid dependence is affected by mGluR-rnediated PI hydrolysis and mGluR-regulated CAMP production.

Key work: opioid; morphine; metabotropic glutamate receptor; dependence; MCPG;

MCCG; MAP4 Introduction

Although opioid analgesics are widely used in the management of pain,

repeated use may lead to the development of tolerance and dependence. Tolerance is

indicated by a decreased efficacy of the drug derchronic use, thereby leading to the

requirement for a higher dose to achieve the desired analgesic efFect. Dependence is a

continued need for the drug to maintain a state of physiological equilibrium, following

repeated administration, and is evidenced by withdrawai symptoms when drug

administration is terminated. Recent evidence supports the involvement of excitatory

amino acid (EAA), or glutamate, receptors in the developrnent of tolerance and

dependence (Fundytus & Coderre, 1994; Marek et ai., 199 1a,b; Trujillo & Akil,

199 1). Specifically, we have previously show that chronic i.c.v. administration of

antagonists selective to metabotropic glutamate receptors (mGluR3s)significantly

attenuates the development of dependence to systemically administered morphine

(Fundytus & Coderre, 1994).

Metabotropic glutamate receptors are directly linked to intracellular second messenger systems via guanine nucleotide regdatory proteins (G proteins; Sladeczek et al., 1985; Sugiyama et al., 1987). Several subtypes of rnGluRs, fiom mGluRl to rnGlu&, have been cloned (Duvoisin et al, 1995; Nakanishi, 1992; Okamoto et al.,

1994; Saugstad, et al, 1994; Schoepp & Conn, 1993). The subtypes of mGluRs can be divided into groups based on receptor pharmacology, signaling cascades and sequence similanties (i-iayashi et al., 1994). The first group of mGluRs consists of mGluRr and mGluR5, which are positively linked to the phosphatidylinosito1 second messenger system (Hayashi et al., 1994; Schoepp Br Conn, 1993). Activation of these receptors leads to phospholipase C (PLC)-mediated phosphatidylinositol (PI) hydrolysis. The second group of mGluRs consists of mGluR2 and mGluR3, which are negatively coupled, via adenylate cyclase, to the productiun of cyclic adenosine-3',5'- monophosphate (CAMP) (Hayashi et al., 1994). The third group of mGluRs consist of mGlu&. 6.7 ,.d 8, which are also negatively coupled to CAMP production, but which show a different receptor pharmacology than mGluRz, in that they are selectively activated by L-amino4-phosphonobutanoate (L-AP4) (Hayashi et al., 1994; Saugstad et al., 1 994).

In a previous study (Fundytus & Coderre, 1994), we showed that chronic intracerebroventricular (i.c.v.) infusion of the mGluR antagonist (S)-4- carboxyphenylglycine ((S)-4C-PG) concurrently with systemic morphine attenuated the development of morphine dependence. Although (S)-4C-PG selectively antagonizes group 1 mGluRs (mGluR, and rnGluRS), it has a secondaiy effect whereby it activates group II rnG1uR.s (mGl&, mGluR,) (Eaton et al., 1993; Hayashi et al.,

1994; Watkins & Collingridge, 1994). It is therefore not entirely clear whether antagonism of mGluRIv5or activation of mGlu& is primady responsible for the ability of (S)-4C-PG to attenuate the developrnent of morphine dependence. To examine the relative contribution of each of the mGluR subtypes, we chose a range of phenylglycine antagonists selective to specific mG1uR.s. We exarnined how non- selective antagonisrn of group 1 (mGluR[ and mGluRs) and group II (mGluR2 and mGluR3) subtypes would affect the development of morphine dependence by adrninistenng a-methyl-4-carboxyphenylglycine (MCPG)(Eaton et al, 1993; Hayashi et al., 1994; Thomsen, Boel & Swdak, 1994) i.c.v. concurrently with systemic morphine. We also selectively antagonized mGluRz and mGl& with 2s, I's,2's-2- methyi-2-(2'-carboxycyclopropyl)-glycine (MCCG) (Jane et al., 1994) and mGl&. ça receptors with a-methyl-L-amino-4-phosphonobutanoate (MAP4) (Jane et al., 1994). respectively. Furthemore, the effects of chronic treatment with these mGluR antagonists on morphine dependence was compared with the effects of acute treatment with the same agents given 10 minutes before naloxone. We now report that chronic non-selective antagonism of rnGluRs with MCPG, and chronic selective antagonisrn of either group II or III mGluRs significantly attenuates the development of morphine de pendence.

Materiais and Methods

Siibjecfs and Surgery

Subjects were male Long Evans hooded rats (Charles River, Quebec) weighing

280-350 grams at the tirne of surgery. Rats were housed in groups of2 to 4, maintained on a 12: 12 hour 1ight:dark cycle (lights on at 06:OO). and given food and water ad libitzim.

On Day O rats were anaethestized with sodium pentobarbital (Somnotol, MTC

Pharmaceuticals, 60 mg kg-'), and a 23 gauge stainiess steel cannula was implanted stereotaxically in the lateral ventrical of each rat (AP = - 1.3 mm and L = - 1.8 mm fiom bregma, and V = -3.8 mm (for chronic infusion) or V = -3 .O mm (for acute i.c.v. injections) fi-om the top of the skull; Paxinos & Watson, 1986). For rats given chronic i.c.v. treatment, the cannula was attached to a Model 2001 Ahet@ osmotic purnp filled with one of the antagonists or vehicle (dilute NaOWsaline). While the rats were still under pentobarbital anaesthesia, one unprimed (not yet pumping) Model 2ML1

Aizet@ osmotic pump containing 50 mg ml-' morphine suKate (gift From Sabex,

Quebec) solution was implanted subcutaneously (M.) on the back of each rat.

Infusion of morphine began approximately 2 to 4 hours following pump implantation.

On the following day, Day 1, rats were bnefly anaesthetized with halothane and a second unprimed Model 2MLl AlzetQ pump containing 70 mg ml-' morphine sulfate solution was implanted S.C. on the back of each rat. Once both pumps were in place, morphine sulfate was continuously infüsed S.C. at a rate of 10 pl hr- 1 from each pump, for a total dose of 36.65 pmol day" (28.8 mg day-'). This two day pump implantation procedure was used to reduce the risk of mortality due to the accumulation of lethal systemic morphine concentrations prior to any tolerance development. Concurrently with morphine, the mGluR antagonists, MCPG (n =20), MCCG (n = 18). MAP4 (n =

17) (Tocris Cookson, Bristol, UK) or vehicle (n = 18) were continuously infused at a rate of 1 pl hr" in a dose of either 1.6, 8 or 40 nmol day-' intracerebroventncularly

(i.c.v.) in rats treated chronically with mGluR antagonists. To assess the effects of chronic administration of the selective mGluR antagonists on general (non-withdrawal) behaviour in rats not dependent on morphine, sorne rats were given 7 days cf i.c.v. vehicle or 40 nmol day-' of either MCPG (n = 3), MCCG (n = 4) or MAP4 (n = 4) without concurrent morphine treatment. The effects of chronic administration of selective mGluR antagonists were also compared with effects of acute administration of the antagonists given as a single i.c.v. injection 10 minutes prior to the induction of

withdrawal. Acute i.c.v. injections of either MCPG (n = S), MCCG (n = 6) or MAP4

(n = 6) were given in a dose of 2 nrnol4 or 4 pl vehicle (n = 15), to rats that

received chronic systemic morphine treatment as descnbed above. A dose of 2 nmol

was chosen to approximate the level received over a 1 to 2 hour period in rats treated

chronically with 40 mol day-'. Doses as high as 99 mol i.c.v. have been used in

leamhg experiments with no obvious side effects (Riedel & Reymann, 1993). The

effects of acute administration of mGluR antagonists on general (non-withdrawal)

behaviour were assessed by obse~ngthe behaviour of non-dependent rats afler a

single i.c.v. injection of either MCPG (n = 4), MCCG (n = 4) or MAP4 (n = 4).

Meamrentent of WfthdrawaIand Non- WilhdrawaI Behavzours

Precipitated abstinence symptoms were assessed on the seventh day of

treatment derinjection of the opioid antagonist naloxone. Naloxone hydrochloride

(Research Biochemicals Inc., Natick, MA) was injected S.C. in a volume of 1 ml kg-'

for a dose of 1 mg kg". In chronically treated rats, behaviour was observed for 10

minutes before and 40 minutes dernaioxone injection, during which time withdrawal

symptoms were assessed by measuring the arnount of time spent teeth chattering and writhing, as well as by counting jumps and wet dog shakes. In rats given an acute i.c.v. injection of mGluR antagonists, withdrawal behaviours were assessed for 10 minutes pnor to i.c.v. injection, 10 minutes after i.c.v. injection but before naloxone, and for 40 minutes after the injection of naloxone. For both chronic and acute conditions, agitation was assessed by rating the severity of vocalization upon light brushing of the back of the neck at both 20 and 40 minutes after the injection of naloxone. Seventy of vocalization was rated on a scale of O to 3 where O = absent and 3 = severe. Also, for both chronic and acute i.c.v. treatment conditions, the time spent in withdrawal and non-withdrawal behaviours (ambulating, rearing, grooming and resting) was also measured for cornparison in non-dependent rats (not given morphine) and morphine- dependent rats (given chronic s.~.morphine).

Statisiical Analyss

Tirned withdrawal behaviours (teeth chattering, writhing) were analyzed using

1 way ANOVA comparing the effects of various doses of each drug with the vehicle control group. Since ail drugs were dissolved in the sarne vehicle, a single control group was used for each experiment to minirnize the number of anirnals exposed to the full-fledged opioid withdrawai syndrome. Testing of the vehicle-treated rats was spread across the testing days for the expenmental anirnals. Significant effects were hrther analyzed using post-hoc LSD t-tests. Counted withdrawal behaviours (number of jumps and wet dog shakes) and severity of agitation were analyzed using a Kniskal-

Wallis ANOVA for non-parametric data, followed by Mann-Whitney U-tests on significant main effects.

The effect of chronic antagonism of subtypes of mGluRs on non-withdrawal behaviours (arnbulating, rearing, groorning and resting) was assessed by comparing the first two time blocks (i-e. 10 minutes prior to naloxone injection and 10 minutes afier naloxone injection) for rats in each treatrnent group. The effects of acute i.c.v. injection of mGluR antagonists was assessed by comparing non-withdrawal and withdrawal behaviours for the first three time blocks (Le. 10 minutes before i.c.v injection, 10 minutes after i.c.v. injection but before naloxone injection, and 10 minutes dernaloxone injection) in non-dependent and morphine-dependent rats. In both cases, a 3-way mixed ANOVA with i.c.v. treatment and morphine treatment as independent variables and time block as a repeated measure was performed on the percent of time spent in each behaviour. Significant effects were further analyzed with post-hoc LSD t-tests.

Results

Figure 1 illustrates the severity of abstinence symptoms during the 40 minute withdrawal penod in rats chronically infused with s.c. morphine and either vehicle,

MCPG, MCCG or MAP4 i.c.v. This expenment was performed to determine if chronic blockade of mGluRs would attenuate the development of morphine dependence. Chronic s.c. administration of 36.65 pmol day'' morphine sulfate resulted in an intense and reliable withdrawal syndrome, evidenced by the occurrence of teeth chattering, wtithing, jumping, wet dog shaking and vocalization on touch (agitation) in vehicle-treated rats.

Figure 1A shows the arnount of time spent in withdrawal (teeth chattering and writhing combined) dunng the 40 minute withdrawal period for morphine-dependent rats treated concurrently with either i.c.v. vehicle, or 1.6, 8 or 40 nmol day" of either

MCPG, MCCG or MAP4. ANOVA indicated a significant effect of MCCG (F(3,32)=

4.14, P < 0.05) and MAP4 (Fc3JI,= 4.62, P < 0.05), but not MCPG @,3,34, = 1.90, P >

0.05). Both MCCG and MAP4 sigruficantly decreased the the spent in withdrawal, as compared to the vehicle control group, at a dose of 1.6 mol day-'. Only MCCG was effective at 8 nmol day-', and only MAP4 was effective at 40 nmol day".

Although there was a trend for the non-selective mGIuR antagonist MCPG to reduce the time spent in withdrawal, this effect failed to reach statistical significance.

Figure 18 illustrates the average frequency of counted symptorns (jumps and wet dog shakes combined) dunng the 40 minute withdrawal penod for morphine- dependent rats. Kruskal-Wallis ANOVA for non-parametric data indicated a significant effect of MCPG (t43.38)= 1 1.74, P < 0.05) and MCCG w3;6) = 14.14, P <

O.OI), but not MAP4 @43,3~1= 6.08, P > 0.05). AS indicated, MCPG significantly decreased the frequency of counted symptoms compared to the vehicle-treated control group at al1 doses used. MCCG significantly decreased the occurrence of counted symptoms at the highest dose, 40 mol day-'. Although there was a trend for MAP4 to decrease the frequency of counted symptoms as dose was increased, this effect failed to reach statistical significance.

Figure 1 C shows the average severity of agitation, as indicated by vocalization upon being lightly touched on the back of the neck, for morphine-dependent rats tested at 20 and 40 minutes post-naloxone. Kniskal-Wallis ANOVA for non- parametric data indicated a significant effect of MCPG @&3,le = 1 1.65, P < 0.05), but not MCCG (w3.36) = 7.64, P > 0.05) nor MAP4 (I-&335) = 3.36. P > 0.05). The non- selective antagonist MCPG significantly decreased the seventy of agitation at al1 doses used. Although MCCG appeared to attenuate the seventy of agitation at 40 nmol day- ', Kruskal-Wallis ANOVA failed to reach statistical significance. MAP4 did not significantly affect agitation at any of the doses used.

To ver@ that chronic infusion of mGluR antagonists had limited effects on general behaviour, non-withdrawal and withdrawal behaviours were compared during the 10 minutes pnor to naloxone injection and the 10 minutes after naioxone injection for non-dependent and morphine-dependent rats chronically infùsed i.c.v. with either vehicle or 40 nmol day-' of MCPG, MCCG or MAP4. Statistics confirmed that rats were more active earlier in the test session, with more time spent ambulating, rearing and grooming and less time spent resting, regardless of i.c.v. treatment (LSD t-test, P

< 0.05). Prior to the injection of naloxone non-dependent and morphine-dependent rats behaved very sirnilarly. Mer the injection of naioxone, non-dependent rats spent more time in non-withdrawal, and less time in withdrawal, behaviours than morphine- dependent rats (LSD t-test, P < 0.05). There were only a few effects of i.c.v. treatment on generai activity level. Regardless of morphine treatment, i.c.v. MCPG- and MAP4-treated rats ambulated more than i.c.v. vehicle-treated rats (P < 0.05, LSD t-test). Also, non-dependent i.c.v. MCCG- and MAP4-treated rats reared more than non-dependent i.c.v. vehicle-treated rats (P < 0.05, LSD t-test). (Data not shown).

Figure 2 depicts withdrawal symptoms during the 40 minute withdrawal period for morphine-dependent rats given an acute i.c.v. injection of either vehicle or 2 nmol of either MCPG, MCCG or MAP4 10 minutes pnor to naloxone injection. This experiment was performed to determine if acute blockade of mGluRs would decrease the expression of abstinence symptoms once dependence had developed. Figure 2A illustrates the time spent in withdrawai (teeth chattenng and writhing combined) during the 40 minute withdrawal period for dependent rats given an acute i.c.v. injection of either vehicle or 2 mol of either MCPG, MCCG or MAP4.

ANOVA indicated a significant effect of i.c.v. treatment (F8.211)= 4.20. P < 0.01).

Acute injection of 2 nmol of MCCG 10 minutes before the precipitation of withdrawal significantly increased the time spent teeth chattering and writhing.

Figure 2B shows the average fiequency of counted symptoms (jumps and wet dog shakes combined) during the 40 minute withdrawai period for dependent rats given an acute icv. injection of either vehicle, MCPG, MCCG or MAP4. Kniskal-

Wallis ANOVA for non-parametric data indicated that there were no differences between vehicle-treated rats and rnGluR antagonist-treated rats (t43.32) = 0-40, P >

0.05).

Figure 2C shows the severity of agitation for dependent rats given an acute i.c.v. injection. Again, Kniskal-Wallis ANOVA for non-pararnetnc data indicated that there were no differences between vehicle-treated and mGluR antagonist-treated rats

(k&3,32) = 1.84, P > 0.05).

To venfy that acute i.c.v. injection of mGluR antagonists had no significant effects on general behaviour, non-withdrawal and withdrawal behaviours were compared dunng the 10 minutes prior to i.c.v. injection, the 10 minutes after i.c.v. injection but before naloxone, and the 10 minutes after naloxone injection in non- dependent and morphine-dependent rats given an acute i. c. v. injection of either vehicle or 2 mol of MCPG, MCCG or MAP4. Statistics confirmed that rats were more active in earlier time blocks (exhibiting more ambulating, rearing and grooming, and less resting) (LSD t-test, P < 0.05), and that &er the injection of naloxone morphine- dependent rats exhibited significantly more time in withdrawal and therefore less time in non-withdrawal behaviours than non-dependent rats (LSD t-test, P < 0.05). The only difference between vehicle-treated and mGluR antagonist-treated rats was that rats given MCCG exhibited significantly more time in withdrawal than vehicle-treated rats after the injection of naloxone (P < 0.05, LSD t-test). (Data not shown).

Discussion

In the present study we have show that chronic antagonism of mGluRs in general, as well as chronic antagonism of group II or IU mGluRs, attenuates the severity of the precipitated morphine withdrawal syndrome. Non-selective antagonism of mGluRs with MCPG decreased the fiequency of jumps and wet dog shakes, as well as the severity of agitation. Selective chronic antagonism of mGl& and mGluR3 with

MCCG significantly decreased the time spent teet h chattering and writhing, the fiequency ofjumps and wet dog shakes, and the severity of agitation. Selective chronic antagonism of mGlu&, 6.7 md 8 with MAP4 decreased the time spent teeth chattering and writhing. Aithough there was also a trend for MAP4 to decrease the fiequency of jumps and wet dog shakes, this effect failed to reach significance. Acute i.c.v. administration of mGluR antagonists just prior to the precipitation of withdrawal failed to decrease the severity of abstinence symptoms, and in the case of MCCG, it actually increased the severity of the withdrawal syndrome. Despite the fact that chronic administration of each antagonist reduced precipitated withdrawal symptoms, we did not. however, observe a clear dose relationship for any of the agents. Perhaps this is due to the fact that Our dose range was relatively smail and the doses may have been in the same effective range.

However, we avoided higher doses to prevent non-selective effects. MCPG has been shown to selectively antagonize group 1 and II mGluRs, with no effects on ionotropic glutamate receptors at concentrations up to 1 mM in vitro (Thomsen et aI, 1994). It has been show that MCCG is selective for group LI mGluRs and MAP4 is selective for group UI mGluRs at concentrations up to 300 pM Ni vitro (Jane et al, 1994), and had no effect on ionotropic receptors at concentrations as high as 500 pM (Jane et al,

1994). Although Our highest infusion (40 nmol day-') is a higher concentration (1667 w),Our two lower doses (1.6 and 8 mol day-') are within the selective concentration range (66.67 and 3 3 3 -3 pM, respectively). Therefore, because the effects were evident at these lower doses, they are likely due to actions at mG1uRs.

There were also differences in the degree to which each antagonist afEected the vanous behaviours. This could possibly be due to inter-subject variability, or possibly because some minor or recessive withdrawai syrnptoms will sometimes increase as major or dominant symptoms decrease.

While chronic i.c.v. antagonism of mGluRs effectively decreased the seventy of the withdrawal syndrome, there were few effects on non-withdrawal behaviours. The only effect due to i.c.v. treatment was that antagonist-treated rats were somewhat more active than vehicle-treated rats. However, the most significant changes in general behaviours were completely independent of treatment given. Thus, rats, including those treated only with vehicle, were generally less active later in the test session because by this time they had explored the test box and were habituated to the environment.

The ability of chronic i.c.v. administration of MCCG and MAP4 to decrease the seventy of morphine withdrawal suggests a role for mûla-regulated CAMP production in the development of opioid dependence. MCCG selectively antagonizes group II mGluRs (mGluR2 and mGluR3), and MAP4 selectively antagonizes group III mGluRs (mGl&,a 7 rad 8), both of which are negatively coupled to CAMPproduction.

There is a high expression of mRNA for group II and III mGluRs, as well as opioid receptors, in thalamus, striatum and cortex (Masu et al, 1994; Mansour et al, 1995).

Because activity at opioid receptors also affects CAMP production, it cm be hypothesized that group U and 111 mGluRs interact with opioid receptors via actions on CAMPproduction. It is generally accepted that chronic opioid treatment leads to compensatory changes in CAMP in neuronal tissues. Thus, whereas acute administration of p- and 6-opioids decreases CAMP production, dunng chronic treatment CAMP production retums to near control levels, and is greatly enhanced during withdrawal (Childers, 199 1). We propose that by antagonizing group II or 111 mGluRs, we are removing one source by which the production of CAMP is decreased, thereby modulating the chronic effects of opioids and possibly eliminating the need for the elicitation of compensatory rnechanisms. As well as antagonizing group II rnGluRs, MCPG also antagonizes group 1 mGluRs (mGluR1 and mGluR5) which are positively coupled to PI hydrolysis. In a previous study (Fundytus & Coderre, 1994) we showed that selective antagonism of group 1 mGluRs with (S)-4C-PG attenuated the severity of morphine withdrawal.

Because activity at opioid receptors also affects PI hydrolysis, these results suggest that mGluR-mediated changes in PI hydrolysis are aiso invoived in the development of opioid dependence. There is a high level of expression of mWA for opioid receptors, as well as group 1 mGluRs, in stnatum and cortex (Masu et ai, 1994; Mansour et al,

1995), suggesting that group 1 mGluRs and opioid receptors rnay interact. There is evidence that while acute administration of p-opioids decreases PI hydrolysis, during chronic administration PI hydrolysis rnay increase to near control levels, and dunng withdrawal it is greatly enhanced (Barg et ai, 1994; Busquets et al, 1995; Dixon et ai, 1990; Narita et al, 1994), suggesting that compensatory mechanisms rnay be elicited during chronic opioid treatment. We propose that by chronically antagonizing group 1 mGluRs, we rnay be decreasing PI hydrolysis and thereby counteracting compensatory increases which rnay be elicited by chronic opioid treatment.

In surnmary, we have demonstrated the involvement of group U and III mGluRs, in addition to our previous evidence for a role of group 1 mGluRs, in the developrnent of opioid dependence. Therefore, treatments which target mGluRYsrnay be valuable tools in decreasing the incidence of opioid dependence. Acknowledgments

This work was supported by MRC gants MT-1 1045 and MT-13236 and a gant from the Stairs Memonal Fund awarded to T.J.C. M.E.F.was supported by an

FCAR studentship. References

BARG,J., BELCHEVA,M.M., ZIMLICHMAN,R., LEVY,R., SAYA,D., MCHALE, R.J.,

JOHNSON,F.E., COSCIA,C. J. & VOGEL,2. (1994). Opioids inhibit endothelin-

mediated DNA synthesis, phosphoinositide turnover, and ~a2+rnobiliiation in rat

C6 glioma cells. J. Neurosci., 14, 5858-5864.

BUSQ~ETS,X.. ESCRIBA, P.V., SASTRE,M. & GARCIA-SEVILLA,J.A. (1995). Loss of

protein kinase C-alpha beta in brain of heroin addicts and morphine-dependent

rats. J. Nezrrochem., 64, 247-252.

CHILDERS.S. R. (1 99 1). Opioid receptor-coupled second rnessenger systems. Llje

SCI., 48, 199 1-2003.

DIXON. W., TING,Y.-W. & CHANG,P.L. (1990). The effects of morphine on

norepinephrine-stimulated phosphatidylinositol response in rat cerebral cortex.

Prog. Ch. Biol. Res.. 328, 279-282.

DUVOISN,R.M., ZHANG,C. & WONELL,K. (1995). Glutamate receptor expressed

in the retina and olfactory bulb. J. Neurosci.. 15, 3075-3083.

EATON,S.A., Jm,D.E., JONES, P.L.ST.J., PORTER,R.H.P., POOL P.C.-K., SUNTER,

D.C., UDVARHELYI,P.M., ROBERTS,P.J., SALT.T.E. & WATKINS,J.C. (1 993).

Competitive antagonism at metabotropic glutamate receptors by (51-4-

carboxyphenylglycine and (0-a-methyl-4-carboxyphenylglycine. Eur. J.

Pharmacol. Mol. Pharmacol., 244, 195 - 197. FUNDYTUS, M.E. & CODERRE,T.J. (1994). Morphine dependence is affected by

activity at metabotropic, as well as ionotropic OA),glutamate receptors. Br.

J. Pharmacol., 113, 1215-1220.

HAYASHI,Y., SEKIY~,N., NAKANISHI, S., JANE,D.E., SUMER, D.C., BIRSE,E.F.,

UDVARKELYI,P.M. & WAWS, J.C. (1994). Analysis of agonist and antagonist

activities of phenylglycine derivatives for dEerent cloned metabotropic glutamate

receptor subtypes. J. Neurosci., 14, 33 70-3377.

JANE, D.E., JONES, P.L.ST.J., POOK, P.C.-K.,TSE, H.-W. AND WATKINS,J.C. (1994).

Actions of two new antagonists showing selectivity for different sub-types of

rnetabotropic glutamate receptor in the neonatal rat spinal cord. Br. J.

Pharmacol., 112, 809-8 16.

MANSOIJR,A., FOX, C.A., Am, H. & WATSON,S.J. (1995). Opioid-receptor rnRNA

expression in the rat CNS: anatomical and functional implications. Trends

Nezrrosci., 18, 22-29.

WK,P., BEN-ELIYAHU, S., GOLD, M. & LIEBESKIND,J-C. (199 la). Excitatory

amino acid antagonists (kynurenic acid and MEC-801) attenuate the development of

morphine tolerance in rats. Bruin Res.. 547, 77-8 1.

WK,P., BEN-ELIYAHU,S., VACCARINO,A.L. & LEBESKIND,J.C. (199 1b).

Delayed application of MK-80 1 attenuates development of morphine toierance in

rats. Brain Res., 558, 163-165. MASU, M., NAKAJIMA,Y., MORIYOSHI,K., ISHII, T., AKAZAWA, C.& NAKANISHI,S.

( 1994). Molecular characterization of NMDA and metabotropic glutamate

receptors. Am. NY Acad Sci., 707, 153- 164.

NAKANISHI,S. (1992). Molecular diversity of glutamate receptors and implications

for brain fùnction. Science, 258, 597-603.

NARITA,M., MAKMUW, M., FENG, Y., HOSKINS,B. & HO, 1.K. ( 1994b). Influence

of chronic morphine treatment on protein kinase C activity: companson with

butorphanol and implication for opioid tolerance. Brain Res., 650, 175- 179.

OKAMOTO,N., HORI, S., AKAZAWA, C., HAYASEII,Y., SHIGEMOTO,R., Mrzmo, N.

& NAI~ANIsHI,S. (1994). Molecular charactenzation of a new metabotropic

glutamate receptor mGluR7 coupled to inhibitory cyclic AMP signal transduction

J. Biol. Chem., 269, 123 1-1236.

PAXMOS, G. & WATSON,C. (1986). The rat brain in stereotaxic coordinates.

Academic Press Inc., San Diego, Califomia.

REDEL, G. & REYMANN, K.(1993). An antagonist of the metabotropic glutamate

receptor prevents LTP in the dentate gyrus of freely moving rats.

Ne~iropharmacol.,32, 929-93 1.

SALT,T.E. & EATON,S.A. (1995). Distinct presynaptic metabotropic receptors for L-

AP4 and CCGI on GABAergic terminais: Phamacologicai evidence suing novel

a-methyl derivative mGluR antagonists, MAP4 and MCCG, in the rat thalamus

in vivo. Netirosci., 65, 5-13. SAUGSTAD,J.A., WE,J.M., MULVMILL, E.R., SEGERSON, T.P. & WESTBROOY

G.L. ( 1994). Cloning and expression of a new member of the L-2-amino-4-

phosphonobutyric acid-sensitive class of metabotropic glutamate receptors. Mol.

Pharmacol., 45, 3 67-3 72.

SCHOEPP,D.D. & CONN, P.J. (1993). Metabotropic glutamate receptors in brain

function and pathology. Trenh Pharmacol. Sci., 14, 13-25.

SLADECZEK,F., PIN,J.P., UCASENS,M., BOCKAERT, J. & WEISS,S. (1985).

Glutamate stimulates inositol phosphate formation in striatal neurones. Natiire.

317, 717-719.

SUGIYAMA,H., ITO, 1. & H~RONO,C. (1987). A new type of glutamate receptor

linked to inositol phospholipid metabolism. Ndure, 325, 53 1-533.

THOMSEN,C., BOEL,E. & SUZDAK,P.D. (1994). Actions of phenylglicine analogs at

subtypes of the meraboiropic glutamate receptor family. Eur. J. Phormacol., 267.

77-84.

TRUJILLO,K.A. & AKn,H. (1991). Inhibition of morphine tolerance and dependence

by the NMDA receptor antagonist MK-80 1. Science, 251, 85-87.

WATKINS,J. C. & COLLNGRIDGE,G. L. (1994). Phenylglycine derivatives as

antagonists of merabotropic glutamate receptors. Trends Pharmacol. Sci., 15, 3 3 3 -

342. Figure 1

A) Mean time spent in withdrawal (teeth chattenng and writhing combined) during the

40 minute withdrawal penod for morphine-dependent rats treated chronically with either vehicle (a),MCPG (C),MCCG (m) or MAP4 (0)i.c.v.

* significantly less than control (P < 0.05, LSD t-test)

B) Average fiequency of counted symptoms (jumps and wet dog shakes combined) during the 40 minute wirhdrawd period for morphine-dependent rats given either vehicle (m), MCPG @), MCCG (C)or MN4 @).

* significantly less than control (P < 0.05, Mann-Whitney U-test)

C) Average severity of agitation dunng withdrawal for morphine-dependent rats given either vehicle (m), MCPG @), MCCG 62)or MM4 @).

* significantly less than control (P < 0.05, Mann-Whitney U-test) 0 ' I T 1 O 1.6 8 40 dose (nmd day -l ,i.c.v.)

O 1.6 8 40 dose (nmol day -l ,i.c.v.)

O 1.6 8 40 dose (nmol day -l , i.c.v.) Figure 2

A) Mean time spent in withdrawal (teeth chattering and writhing combined) during the 40 minute withdrawa! for morphine-dependent rats given an acute i.c.v. injection of either vehicle, or 2 nrnol of MCPG, MCCG or MAP4 10 minutes prior to the precipitation of withdrawal.

# significantly greater than vehicle (LSD t-test, P < 0.05)

B) Average frequency of counted symptoms (jumps and wet dog shakes combined) during the 40 minute withdrawal period for morphine-dependent rats given an acute i.c.v. injection of either vehicle, MCPG, MCCG or MAP4 just prior to the precipitation of withdrawal.

C) Average severity of agitation for morphine-dependent rats given an acute i.c.v. injection of either vehicle, MCPG, MCCG or MAP4. - ve.hit.de MCPG MCCG MAP4 i.c.v. treatment

i.c.v. treatment CHAPTER 4: Assessment of the relationship behveen products of

phospbatidylinositol hydrolysis and morphine dependence In the previous two chapters, we examined the effects of selective antagonism of mGluR subtypes on the development of morphine dependence. Chronic antagonism of rnGluRs positively coupled to PI hydrolysis, as well as mGluRs negatively coupled to CAMPproduction, significantly attenuated the development of morphine dependence.

The relationship between CAMPand opioid tolerance and dependence has been thoroughly investigated both in vitro and in vivo (Childers, 199 1). Conversely, although there is a fair amount of evidence suggesting a relationship between PI hydrolysis and acute and chronic administration of opioids, very little has been done to examine how altering PI hydrolysis may affect the development of tolerance and dependence.

Recent work has shown a clear relationship between glutamate, opioids and PI hydrolysis. First, opioids affect glutamate-activated currents in a manner similar to

PKC, a product of PI hydrolysis (Chen and Huang, 1991). Moreover, morphine analgesia has been shown to be afFected by drugs which decrease PKC (Raffa et al,

1992; Raffa and Martinez, 1992; Zhang et al, 1990). Several studies have shown that acute administration of p-opioids, as well as administration of opioids in systems in which p-opioid receptors are predominant, induces a decrease in PI hydrolysis (Barg et al, 1994; Barg et al, 1992; Johnson et al, 1994). Dunng chronic administration of p-opioids, some investigators have observed an increase in PI hydrolysis to near control levels (Pellegrini-Giampietro et al, 1988; Nanta et al, 1994b). There is general consensus that during withdrawal from opioids, PI hydrolysis is greatly increased

(Busquets et al, 1995; Pe1egrin.i-Giampietro et al, 1988).

Recently, it has been suggested that inhibition of PKC, a product of PI hydrolysis, may attenuate the developrnent of opioid tolerance and dependence. 1.c.v. infusion of H-7,a non-selective protein kinase inhibitor (Hidaka et al, 1984), concurrently with i.c.v. morphine or butorphanol, attenuated the developrnent of analgesic tolerance to i.c. v. administered morphine or butorphanol (Maldonado et al,

1995; Narita et al, 1994a). Although H-7 inhibits PKC, it also inhibits PKA and cGMP-dependent protein kinase at similar concentrations (Hidaka et al, 1984). Along with H-7, it has also been show that GMI ganglioside attenuates the development of tolerance and dependence when CO-administeredwith intrathecal (i.t.) morphine (Mao el al, 1994; Mayer et al, 1995). However, as with H-7,although GM 1 gangiioside does inhibit the translocation of PKC, its effects are not selective. GM 1 ganglioside also inhibits the activation of phospholipase Az (Hungund et al, 1994), and modulates the activity of ca2- channels (Carlson et al, 1994; Bressler et al, 1994). Thus, because

H-7 and GM 1 ganglioside are not selective, it is not entirely clear that their ability to decrease the development of tolerance and dependence is due to effects on PKC.

Therefore, in the following study, we used a highly selective inhibitor of PKC activation, chelerythrine, to assess the role of PKC in opioid dependence.

C helerythnne inhibits the activation PKC by interacting with its catalytic domain following the translocation of PKC to the ce11 membrane (Herbert, Augereau, Gleye and Maffrand, 1990). PI hydrolysis also leads to intracellular ca2' release via production of IF3, which stimulates the release of ca2' fiom stores on the endoplasmic

reticulum. Thus, to further assess the role of PI hydrolysis in opioid dependence, we

examined the effect of inhibition of intracellular ca2+release on morphine withdrawal.

We inhibited the release of intracellular ca2+with thapsigargin, which initially induces

a release of ca2*from intemal stores on the endoplasrnic reticulum, but prevents re-

uptake. thereby inhibiting further ca2+release when given chronically (Thastrup,

Cullen. Drobak, Haniey and Dawson, 1990).

As before, we chronically infused our test agents (in this case chelerythrine and

thapsigargin) i.c.v concurrently with morphine treatment and measured the severity of

precipitated abstinence symptoms on the seventh day of treatment. Our assumption

was that if alterations in PI hydrolysis are involved in the development of morphine

dependence, we should see a reduction in abstinence symptoms in chelerythnne- or

t hapsigargin-treated rats.

Using these highly selective inhibitors of products of PI hydrolysis. the following manuscript supports the involvement of PI hydrolysis in the development of morphine dependence. Selective chronic inhibition of PKC with chelerythrine significantly decreased the severity of precipitated withdrawal symptoms. As well as

PKC inhibition, we also showed that chronic inhibition of intracellular ca2' release wit h thapsigargin significantly attenuated the precipitated morphine withdrawal syndrome. Ours is the first study to demonstrate that highly selective inhibitors of either PKC or intracellular ca2' release attenuate the development of morphine dependence. 2+ Chronic inhibition of intraceliular Ca -release or PKC activation significantly

reduces the development of morphine dependence.

1-23 Marian E. ~undytus~~and Terence J. Coderre

1 2 Pain Mechanisms Laboratory, Clinical Research Institute of Montreal, Department of 3 Psychology, McGill University and Centre de recherche en sciences neurologiques et département de médecine, Université de Montréal.

To whom correspondence should be sent:

Dr. Terence J. Coderre Pain Mechanisms Laboratory Clinical Research Institute of Montreal 1 10 Pine Avenue West Montreal, Quebec, Canada H2W 1R7

Telephone: (5 14) 987-5750 FAX: (5 14) 987-5624 Abstract

We have previously shown that chronic antagonisrn of metabotropic glutamate receptors in the brain attenuates naloxone-precipitated withdrawal syrnptoms in rats treated chronically with subcutaneous (s.c.) morphine. Several subtypes of metabotropic glutamate receptors are directly Iinked, through a guanine nucleotide regulatory protein, to the phosphatidylinositol (PI) second messenger system. In the present investigation, we assessed the effect of inhibiting the products of PI hydrolysis on the development of opioid dependence. Thus, concurrently with subcutaneous morphine, we infùsed intracerebroventricularly (i.c.v.) in rats, various doses of chelerythrine, which selectively inhibits the activation of protein kinase C, and thapsigargin, which inhibits the release of intracellular calcium when given chronically.

Both chelerythrine and thapsigargin reduced the severity of naloxone-precipitated abstinence symptoms when infused i.c.v. at a dose of IO nmol/day. A single injection of either chelerythrine or t hapsigargin immediately pnor to the precipitation of withdrawal failed to decrease the severity of abstinence symptoms. Our results suggest that by chronically inhibiting activity of the phosphatidylinositol system, the development of morphine dependence can be attenuated.

Key Words: opioid withdrawal, metabotropic glutamate receptor, protein kinase C,

inositol- l,4,5-trisp hosphate, chelerythrine, thapsigargin. 1. Introduction

Chronic treatment with opioid analgesics such as morphine leads to the development of tolerance and dependence. Tolerance is a decreased sensitivity to the effects of the dmgs, leading to the requirernent for a higher dose to achieve the desired analgesic effect. Dependence is a continued need for the dmg, following repeated administration, to maintain a state of physiological equilibrium, and results in an aversive withdrawal syndrome upon removal of the dmg. Several investigators have show that concurrent treatment with N-rnethyl-D-aspartate (NMDA) receptor antagonists attenuates the development of both tolerance to and dependence upon morphine (Marek et al, 199 1a; Marek et al, 199 1b; Trujillo and Akil, 199 1 ). NMDA receptors are activated by the excitatory arnino acid (EAA) glutamate, which also acts at two other types of ionotropic receptors (AMPA and kainate) and at rnetabotropic glutamate receptors (mGluR's) (Mayer and Westbrook, 1987; Monaghan et al, 1989).

We have previously show that chronic antagonism of not only NMDA receptors, but also metabotropic glutamate receptors (mGluRs), in the brain attenuates naloxone-precipitated withdrawal symptoms in rats given chronic subcutaneous morphine (Fundytus and Coderre, 1994). Metabotropic glutamate receptors are coupled directly to the cell membrane by a guanine nucleotide regdatory (G) protein

(Sladeczek et al. 1985; Sugiyama et al, 1987). Activation of severai subtypes of the mGluR (mGluR 1a, mGluR 1P and mGluR5) stimulates a phospholipase C catalyzed phosphatidylinositol (PI) hydrolysis (Schoepp and Conn, 1993). Phosphatidylinositol-

4,s-bisphosphate (PIP,) is hydrolyzed to produce diacylglycerol @AG) and inositol- 2+ 1.4,s-trisphosphate (IP,). IP3 promotes the release of Ca fiorn interna1 stores on the endoplasmic reticulum. DAG promotes the translocation and activation of protein kinase C (PKC) (Kapcala et al, 1992).

Both acute and chronic administration of opioids has been show to affect PI hydrolysis. Although there are sorne confiicting results (Smart et al, 1994), many investigators find that acute administration of selective p-opioid agonists leads to a decrease in PI hydrolysis (Barg et al, 1994; Barg et al, 1992; Johnson et al, 1994).

With chronic administration of opioids, PI hydrolysis or PKC activity has been found to be either decreased (Busquets et al, 1995; Pellegrini-Giampietro et al. l988), unchanged (Dixon et al, 1990), or increased (Narita et al, 1994b) compared to untreated controls. However, during opioid withdrawal, PI hydrolysis or PKC activity has been found to be greatly enhanced (Busquets et al, 1995; Mao et al, 1995;

Pellegrini-Giampietro et al, 1988). There is also evidence suggesting that administration of opioids can affect glutamate-receptor mediated activity in a marner similar to the products of PI hydrolysis. It has been shown that application of opioids can rnimic the effects of PKC hy enhancing glutamate-activated currents (Chen and

Huang, 1991).

In addition to application of opioids afFecting PI hydrolysis, it has also been demonstrated that changes in PI hydrolysis and PKC activity cm affect both endogenous and exogenous opioid activity. PKC interacts with endogenous opioids by stimulating secretion of both P-endorphin and it's precursor, pro-opiomelancortin

(Abou-Samira et al, 1987; Kapcala et al, 1992). Moreover, activity of the PI system affects the analgesic efficacy of opioid dmgs (Raffa et al, 1992; RafEa and Mutinez,

19%; Zhang et al, 1990).

The present study was performed to assess whether selective chronic inhibition

2- of PI-stimulated Ca release or PKC activation in the brain could attenuate the severity of the precipitated morphine withdrawal syndrome in rats. This was assessed by chronically infusing specific inhibitors of these intracellular effects

2 - intracerebroventricularly (i.c.v.). Thapsigargin initially stimulates release of Ca i, but

2, prevents re-uptake into the endoplasmic reticulum, thereby inhibiting further Ca release when administered chronically (Thastrup et al, 1990). Chelerythrine selectively inhibits the activation of PKC by interacting directly with its catalytic domain (Herbert et al, 1990). Both thapsigargin and chelerythrine have been show to be both cell permeable (Thastrup et al, 1990; Herbert et al, 1990) and active in vivo (Marks et al,

199 1 ; Perchellet et al, 1993; Young et al, 1994). We also assessed the effects of a single i.c.v. injection of chelerythrine and thapsigargin immediately pnor to the precipitation of withdrawal. The present study shows that chronic, but not acute.

2- inhibition of either Ca release or activation of PKC attenuates the precipitated withdrawal syndrome in rats treated chronically with subcutaneous morphine.

2. Materials and Methods

2. I Subjects and Swgeiy

Subjects were male Long Evans rats (280-3 50 gram; Charles River, PQ). The rats were housed 2 to 4 per cage, on a 12: 12 hour lightdark cycle (lights on at 06:OO). with food and water avaiIable ad libitum. On Day O rats were anaesthetized with sodium pentobarbital (Somnotol, MTC

Pharmaceuticals, 60 mgkg), and a 23 gauge staidess steel cannula was implanted stereotaxicdly in the lateral ventncal of each rat (AP = - 1.3 mm and L = - 1.8 mm fiom

bregma, and V = -3.8 mm fkom the top of the skull; Paxinos and Watson, 1986). For rats given vehicle, chelerythrine or thapsigargin chronically, the cannula was attached to a Model 2001 Ahet@ osmotic mini pump fiiled with either chelerythnne, thapsigargin or vehicle (10% DMSO in saline). While the rats were still under pentobarbital anaesthesia, one unpnmed (i.e. not yet pumping) Model 2ML 1 AlzetQ pump containing 60 mglrnl morphine sulfate solution was implanted subcutaneously

(s.c. ) on the back of each rat. Infusion of morphine began approximately 2 to 4 hours following implantation. On the following day, Day 1, rats were bnefly anaesthetized with halothane and a second unprimed Model 2MLl pump containing 60 mghl rnorphine sulfate solution was implanted S.C. on the back of each rat. This two day pump implantation procedure was used to reduce the risk of mortality due to the accumulation of lethal systemic morphine concentrations pnor to any tolerance development. To assess the effects of chronic inhibition of intracellular messengers on behaviour in rats not dependent on morphine, some rats were given i.c.v. vehicle or 10 nmoVday of either thapsigargin or chelerythnne without concurrent morphine treatment (non-dependent). The effects of acute administration of chelerythnne and thapsigargin were assessed by observing the behaviour of non-dependent rats after a single injection of either 1 mol chelerythe or 0.5 nmol thapsigargin. 2.2 Dmgs

Chelerythrine (Calbiochern, San Diego, CA) and thapsigargin (Research

Biochernicals Inc., Natick, MA) were continuously infused i.c.v. at a rate of 1 pVhr in the following doses: 0.1, 1 or 10 nmol/day. Acute injections were given i.c.v. 10 minutes prior to the precipitation of withdrawal in doses of O. 1 or 1 mol chelerythrine, or 0.05 or 0.5 nmol thapsigargin. These doses of chelerythrine and thapsigargin were chosen to approximate the level received over a 1 to 2 hour penod in chronically-treated rats; higher doses were found to produce side effects such as seizures. Morphine sulfate (Sabex, Montreal, PQ) was continuously delivered S.C. at a rate of 10 pVhr, for a total dose of 36.65 pmoUday (28.8 mglday).

2.3 Withdrawai Measurement

Precipitated abstinence symptoms were assessed on the seventh day of morphine treatrnent after injection of the opioid antagonist naloxone (1 mgkg, s.c.).

Rats given vehicle, chelerythrine or thapsigargin chronically were observeci for 10 minutes before and 40 minutes after naloxone injection. Withdrawai symptoms were assessed by measuring the arnount of tirne spent teeth chattenng and writhing, as well as by counting jumps and wet dog shakes. The time spent in non-withdrawal behaviours (ambulatin& rearing, grooming and resting) was also measured. The time spent in withdrawal and non-withdrawal behaviours was also measured for cornparison, for 10 minutes before and der the injection of naloxone, in non- dependent rats (not given morphine) infused chronically with either vehicle or 10 nmoVday chelerythrine or thapsigargin. Rats given an acute i.c.v. injection of vehicle. chelerythrine or thapsigargin were observed for 10 minutes pior to i.c. v. injection,

then another IO minutes after i.c.v. injection but before naloxone injection, and for 40

minutes following naloxone injection, du~gwhich withdrawal symptoms were

assessed. In rats given an acute i.c.v. injection of vehicle, 1 mol chelerythrine or 0.5

mol t hapsigargin, non-withdrawai and withdrawai behaviours were compared during

the 10 minutes prior to i.c.v. injection, the 10 minutes deri.c.v. injection but before

naloxone injection, and 10 minutes after naloxone injection in both non-dependent and

morphine-dependent rats.

2.4 Statistical Analysis

Timed withdrawal behaviours (teeth chattering, writhing) were anaiyzed using

planned cornparisons between experimental groups and the vehicle control group.

Counted withdrawal behaviours (number ofjumps and wet dog shakes) were analyzed

using a Kruskal-Wallis ANOVA for non-parametric data, followed by Mann-Whitney

U-tests on significant main effects.

The effect of chronic inhibition of release and PKC activation on non- withdrawal behaviours (ambulating, rearing, groorning and resting) was assessed by comparing the first two time blocks (i.e. 10 minutes prior to naioxone injection and 10 minutes aiter naloxone injection) for rats in each treatment group. The effects of acute i.c.v. injection of chelerythnne and thapsigargin was assessed by comparing non- withdrawal and withdrawal behaviours for the first three time blocks (i.e. 10 minutes before i.c.v. injection, 10 minutes after i.c.v. but before naloxone injection, and 10 minutes after naloxone injection). In both cases, a 3-way mixed ANOVA with i.c.v. treatment and morphine treatment as independent variables and time block as a repeated measure was performed on the percent of time spent in each behaviour.

Significant effects were further assessed with Tukey's post-hoc tests for samples of unequa1 sizes.

3. Results

Figure 1 illustrates the severity of abstinence symptoms observed during the 40 minute withdrawal period for morphine-dependent rats treated chronically with i.c.v. vehicle, chelerythrine or thapsigargin. Chronic S.C.administration of 36.65 prnoVday of morphine sulfate produced an intense and reliable withdrawal syndrome, evidenced by the occurrence of teeth chattering, writhing, jumping and wet dog shaking.

Figure 1A shows the amount of time spent in withdrawal (teeth chattering and writhing combined) during the 40 minute withdrawal period for morphine-dependent rats treated concurrently with vehicle, or 0.1, 1 or 10 nmoVday thapsigargin or chelerythrine i.c.v. 90th chelerythrine and thapsigargin produced a reduction in the amount of time spent in withdrawal. The highest dose of each agent, 10 nrnoüday, significantly decreased the amount of tirne spent in withdrawal.

Figure 1B shows the frequency of counted symptoms (jumps and wet dog shakes combined) during the 40 minute withdrawal penod for morphine-dependent rats in each chronic i. C.V. treatment group. AIthough chelerythrine appears to slightly increase the number ofjumps and wet dog shakes, there were no significant differences in the frequency of counted symptorns between any of the treatrnent groups. Figure 2 illustrates the percent of time spent in non-withdrawai (ambulating,

rearing, groorning and resting) and withdrawal (teeth chattering and writhing) behaviours for both non-dependent and morphine-dependent rats treated chronically with i.c.v. vehicle, or 10 nmoVday of either chelerythrine or thapsigargin i.c.v., during the 10 minutes prior to naloxone injection (A and B), and dunng the 10 minutes afler naloxone injection (C and D). Statistics confimed that, regardless of i.c.v. treatment. rats were more active prier to the injection of naloxone, when they spent more time ambulating, rearing and grooming, than afler naloxone, when they spent more time resting. Morphine-dependent rats were generally less active than non-dependent rats. and as expected showed more withdrawal behaviours after the injection of naloxone.

Specific cornpansons at each time period showed that there were no differences between i.c.v. treatment groups pnor to the injection of naloxone for either non- dependent or morphine-dependent rats, or for non-dependent rats der the injection of naloxone (P > 0.05). After the injection of naloxone, morphine-dependent rats given chelerythrine or thapsigargin spent more time in non-withdrawai behaviours (P < 0.05) and less time in withdrswal than morphine-dependent rats given vehicle (P < 0.05; C and D).

Figure 3 illustrates the severity of abstinence symptoms dunng the 40 minute withdrawal period for rats treated chronically with s.c. morphine, and given a single i.c.v. injection of vehicle, chelerythrine or thapsigargin 10 minutes prior to the precipitation of withdrawal. As before, chronic S.C. morphine sulfate (36.65 p moVday) produced an intense and reliable abstinence syndrome evidenced by the occurrence of teeth chattenng, writhing, jumping and wet dog shaking.

Figure 3A shows the arnount of time spent in withdrawal (teeth chattenng and writhing combined) dunng the 40 minute withdrawal period for morphine-dependent rats given a single i.c.v. injection of either vehicle, 0.1 or 1 mol chelerythrine, or 0.05 or 0.5 nmol thapsigargin. Although 1 nmol of chelerythrine appeared to decrease the amount of time spent in withdrawal, the results failed to reach significance.

Figure 3B shows the frequency of counted symptoms (jumps and wet dog shakes combined) during the 40 minute withdrawai penod for morphine-dependent rats given a single i.c.v. injection of vehicle, chelerythrine or thapsigargin. The frequency of counted symptoms was sigruficantly increased by I mol of chelerythrine.

Figure 4 illustrates the percent of tirne spent in non-withdrawal and withdrawal behaviours for both non-dependent and morphine-dependent rats given a single i.c. v. injection of either vehicle, 1 nmol chelerythrine or 0.5 mol thapsigargin, during the IO minutes pnor to i.c.v. injection (A and B), the 10 minutes after i.c.v. injection but before naloxone injection (C and D), and the 10 minutes after naloxone injection (E and F). Statistics coniïrmed that, again, regardless of i.c.v. treatment, rats were more active earlier in the test session, with more time spent arnbulating, rearing and groorning and less time spent resting, than later in the test session. Morphine- dependent rats were generally less active than non-dependent rats, and exhibited more withdrawal symptoms after naloxone injection. There were no differences on any behaviours between i.c.v. treatment groups, except that thapsigargin-treated rats arnbulated less than vehicle-treated rats (P < 0.05).

4. Discussion

The present results suggest that PI hydrolysis contnbutes significantly to the development of dependence with chronic opioid use. Chronic inhibition of products of

PI hydrolysis (IF',-stirnulated c:', release or DAG mediated PKC activation) concurrently with morphine treatment sigxufïcantly decreased the severity of timed withdrawal symptoms, while having very little effect on non-withdrawal behaviours.

Both chelerythrine and thapsigargh, at the highest dose used, 10 nmoVday, produced a significant reduction in timed withdrawal behaviours. A single i.c.v. injection of either chelerythrine or thapsigargin was ineffective in reducing withdrawal behaviours.

Although 1 nmol of chelerythrine appeared to decrease tirne spent in withdrawal. the results failed to reach significance. Furthermore, t his dose of chelerythrine unexpectedly increased the frequency of counted symptoms.

In both acute and chronic experiments, rats spent more time arnbulating, rearing and grooming early in the test session. As the test session progressed, the time spent resting increased and activity decreased, most likely because by this time they had explored the test box thoroughly and were habituated to the environment. This finding is cornmon in behavioural observations. Furthemore, afker the injection of naloxone, morphine-dependent rats in both acute and chronic experirnents spent more time in withdrawal and less time in non-withdrawal behaviours than non-dependent, as one might expect. For both acute and chronically treated rats, there were no differences between i.c.v. treatment groups, with a few exceptions. Rats given an acute i.c.v. injection of 0.5 nrnol thapsigargin arnbulated less than rats given an acute i.c.v. injection of vehicle. Chronic i.c.v. treatment with either chelerythnne or thapsigargin also decreased the time spent in withdrawal (and therefore increased the time spent in non-withdrawal behavioun) cornpared to chronic vehicle-treated rats afler the injection of naloxone.

Recently, experimental findings suggest that there are interactive effects between opioids, glutamate and the products of PI hydrolysis (e-g. PKC and LP,). It has been demonstrated in cultures of spinal trigeminal neurons in thin medullary slices

2 4 5 from rats that acute application of the p-opioid agonist D-Ala -MePhe -Gly-01 - enkephalin (DAMGO) potentiates glutamate activated currents (Chen and Huang,

199 1). This effect of DAMGO seemed to involve PKC, as administration of PKC produced similar effects, and addition of a PKC inhibitor attenuated the current enhancing effects of both PKC and DAMGO (Chen and Huang, 199 1).

Activation of the PI system also affects endogenous opioid systems. It has recently been show that PKC activators can stimulate secretion of the endogenous opioid P-endorphin from cultured hypothalamic cells (Kapcala et al, 1992).

Furthemore, the secretion of pro-opiomelanocortin (POMC), from which P-endorphin is formed, can be stimulated in the anterior pituitary by activating PKC (Abou-Samira et al, 1987).

Moreover, alterations in PI hydrolysis have been demonstrated to affect opioid analgesia. Pre-treatment of mice with lithium chloride reduced the antinociceptive action in the tail-flick test of several p-opioid agonists: morphine, DAMGO and sufentanii (Raffa et al, 1992; Raffa and Martinez, 1992). Lithium chioride inhibits the activity of the enzyme inositol 1-phosphatase, thereby biocking resynthesis of PiP,- from inositol phosphate (ü?)and decreasing PI hydroiysis (Li et al, 1993; Manji et al,

!993). Raffa and colleagues also found that treatment with iP,, which is produced by

PI hydrolysis, restored the efficacy of the opioid anaigesics in rnice pre-treated with lithium chlonde. These results suggest that opioid analgesia may be mediated at least in part by increased PI hydrolysis. However, other investigators (Zhang et al, 1990) observed that activating PKC, a product of PI hydrolysis, attenuated analgesia produced by p-. 6- and K-opioid agonists. Thus, the interaction between opioids and the PI system appears to be a complicated one.

Many investigators have rneasured PI hydrolysis following opioid administration. When using systems in which p-apioid receptors are predominant, or when selective p-opioid agonists are used, a decrease in PI hydrolysis is generally observed upon acute administration of opioid agonists (Barg et al, 1994; Barg et al,

1992; Johnson et al, 1994; but also see Smart et al 1994). Conversely, in systems which are mediated predominantiy by 6- or K-opioid receptors, or when selective 6- or

K-opioid agonists are used, an enhancement of PI hydrolysis is generally observed

(Barg Cr al, 1993; Feng et ai, 1994; Jin et al, 1994; Leach et al, 1986; Okajima et al,

1993; Penyasarny and Hoss, 1990; Smart et al, 1994; Tsu et al, 1995; but see also Yu and Sadee 1986). Recent studies have also examined the effects of chronic opioid treatment on

PI hydrolysis. Hydrolysis has been measured during both the tolerant/dependent state

(prior to the precipitation of withdrawal) and during the withdrawal state. There is some disagreement as to the activity of the PI system during the toleraddependent state. Dixon et al (1990) found decreased PI hydrolysis, as indicated by decreased levels of P,P, (inositol bisphosphate) and IP,, in cerebrai cortices taken from rats that had been treated continuously with morphine for 24 hours, as compared to an untreated control group. They also found a decrease in the norepinephrine-stimulated accumulation of the products of PI hydrolysis. Similarly, Busquets et al (1 995) observed that in brains taken fiorn human heroin addicts that had died of an overdose,

PI hydrolysis was decreased, as indicated by a decreased level of PKC-aP in frontal cortex. Moreover, these same investigators found that in dependent rats treated chronically with morphine, PI hydrolysis was decreased as indicated by decreased levels of PKC-aP in fiontal cortex cornpared to untreated controls, while in chronic morphine-treated rats undergoing naloxone-precipitated withdrawal, there were increased levels of PKC-aP. Conversely, Pellegrini-Giampietro et al (1 988) observed that in cortical slices taken from dependent rats (cultured in medium containing morphine) norepinephrine- and carbachol-induced PI hydrolysis, measured by levels of inositol phosphates, was not different from that observed in cortical slices taken fiom non-dependent rats. Under withdrawal conditions (where the morphine in the medium is replaced by naloxone) the norepinephrine- and carbachol-induced PI hydrolysis and accumulation of inositol phosphates was greatly enhanced (Pellegnni-Giampietro et al, l988), sirnilar to the withdrawal effects observed by Busquets et al (1 995). Narita el al ( 1994b) found increased PI hydrolysis, as indicated by increased PKC activity, in rats treated chronically with morphine. In these studies, PI hydrolysis after chronic opiate treatment was compared only to untreated controls, and not to rats or tissues which had been given a single (acute) opiate treatment. If acute opioid treatment produces a highly significant decrease in PI hydrolysis, then in chronic opioid-treated rats PI levels which are slightly lower or stightly greater than control vaiues may in fact represent a compensatory increase in activity of the PI system, as seen in the studies conducted by Pellegrini-Giampietro et al (1988) and Narita et al (1 994b).

Thus, although these results are contradictory with respect to the activation state of the PI system during the tolerant/dependent state, they each suggest that there are significant effects of chronic opioid administration on PI hydrolysis. Moreover, they suggest that during chronic morphine treatment compensatory mechanisms may be induced in the central nervous system, with a resultant over-compensation during the withdrawal state.

Recently, it has been shown that inhibiting protein kinases can attenuate the development of tolerance to i.c.v opioids. Lc.v. Infusion of H-7, a non-selective protein kinase inhibitor concurrently with i.c.v. morphine or butorphanol, attenuated the development of analgesic tolerance to i.c.v. administered morphine or butorphanol

(Narita, et al, 1994a).

Recent evidence suggests that chronic opioid treatment may elicit compensatory changes in the spinal cord as weU as the brain. Mao et al (1995) found that although an acute intrathecal administration of morphine had no effect on the nurnber of PKCy immunostained neurons in the spinal cord dorsal hom of rats, chronic intrathecal morphine increased the number of immunostained neurons. Furthemore, It has previously been shown that coadmuiistration of GM1 ganglioside with intrathecal morphine attenuates the development of tolerance and dependence (Mao et al, 1994;

Mayer et al, 1995). Although GMI ganglioside inhibits the translocation of PKC, its effects are not selective. GMI ganglioside aiso inhibits activation of phospholipase A2

(Hungund et al, 1994). as well as modulating the activity of calcium channels (Carlson ef al, 1994; Bressler et al, 1994). Thus, it is not entirely clear that the effects of GM 1 ganglioside in these experiments are necessarily due to its inhibition of the translocation of PKC.

2 - Our results indicate that chronically inhibiting either the IPptimulated Ca release or the activation of PKC in the brain significantly reduces the development of morphine dependence. We suggest that opioid-receptor activation does indeed lead to decreased PI hydrolysis, with a compensatory increase in the activity of the PI system

2- during chronic opioid treatment. Thus by inhibiting Ca release or the activation of

PKC in the brain, both eRects of PI hydrolysis, we were able to counteract the elicitation of a cornpensatory increase in activity of the PI system, thereby reducing the effects of PI hydrolysis on morphine dependence. Acknowledgements

This work was supported by MRC grant #MT- 1 1045 and a gant from the

Stairs Mernorial Fund awarded to T.J-C. M.E.F. was supported by an FCAR studentship. References

Abou-Samira, A.B., J.P. Harwood, K.J. Catt and G. Aguilera, 1987, Mechanisms of

action of CRF and other regdators of ACTH reIease in pituitary corticotrophs.

Ann. NY Acad. Sci., 512, 67.

Barg, J.. M.M. Belcheva and C.J. Coscia, 1992, Evidence for the implication of

phosphoinositol signal transduction in mu-opioid inhibition of DNA synthesis, J.

Neurochem. 59, 1 145.

Barg, J., M.M. Belcheva, R. Zimlichman, R. Levy, D. Saya, R.J. McHale, F.E.

Johnson, C.J.Coscia and 2. Vogel, 1994, Opioids inhibit endothelin-mediated

DNA synthesis, phosphoinositide turnover, and ~a2+mobilization in rat C6

gliorna cells, J. Neurosci. 14, 5858.

Barg, J., S.Y. Nah, R. Levy, D. Saya and Z. Vogel, 1993, Modulation of thymidine

incorporation by kappa-opioid ligands in rat spinal cord-dorsal root ganglion CO-

cultures. Brain Res. 629, 109.

Bressler, J.P., L. Belloni-Olivi and S. Forman, 1994, Effect of ganglioside GMl of

arachidonic acid release in bovine aortic endothelial cells, Life Sci. 54, 49.

Busquets, X., P.V. Escnba, M. Sastre and J.A. Garcia-Sevilla, 1995, Loss of protein

kinase C-alpha beta in brain of heroin addicts and morphine-dependent rats, J.

Neurochem. 64, 247.

Carlson, R.O.,D. Masco, G. Grooker and S. Spiegel, 1994, Endogenous ganglioside

GM 1 modulates L-type calcium chamel activity in N 18 neuroblastoma celfs, J.

Neurosci. 14, 2272. Chen, L., and L.-Y.M. Huang, 1991, Sustained potentiation of NMDA receptor-

mediated glutamate responses through activation of protein kinase C by a 1 opioid,

Neuron 7, 3 19.

Dixon, W., Y.-W.Ting and P.L. Chang, 1990, The effeas of morphine on

norepinephrine-stimulated phosphatidylinositol response in rat cerebral cortex;

Prog. Clin. Biol. Res. 328, 279.

Feng, Y.Z., M. Narita, M. Makimura, B. Hoskins and I.K. Ho, 1994, Enhancement of

the rate of GTP hydrolysis in rat brain by repeated kappa-opioid treatment, Brain

Res. 667, 133.

Fundytus, M.E. and T.J. Coderre, 1994, Morphine dependence is affected by activity

at metabotropic, as well as ionotropic (NMDA), glutamate receptors, Br. J.

Pharmacol. 113, 1215.

Herben, LM., J.M. Augereau, J. Gleye and J.P. Maffrand, 1990, Chelerythrine is a

potent and specific inhibitor of protein kinase C, Biochem. Biophys. Res.

Commun. 172, 993,

Hungund, B.L., A. Zheng, L. Lin and A.I. Barkai, 1994, Ganglioside GM 1 reduces

ethanol induced phospholipase A2 activity in synaptosomal preparations from

rnice, Neurochem. Intematl. 25, 32 1.

Jin, W., N.M. Lee, H.H. Loh aiid S.A. Thayer, 1994, Opioids mobilize calcium from

inositol 1.4,s-trisphosphate-sensitive stores in NG108- 15 cells, J. Neurosci. 14.

1920. Johnson, P.S., J.B. Wang, W.F. Wang and G.R. Uhl, 1994, Expressed mu opiate

receptor couples to adenylate cyclase and phosphatidylinositol turnover,

NeuroReport 5, 507.

Kapcala, L.P., C.-F. Wang and H.-H. luang, 1992, Protein kinase C activators

stimulate beta-endorphin secretion fiom hypothalamic cells, Brain Res. Bull. 29,

553.

Laschka, E., H.J. Teschemacher, P. Mehraein and A. Herz, 1976, Sites of action of

morphine involved in the development of physical dependence in rats,

Psychophamacologia (Berl.) 46, 14 1.

Leach, R.P., S.B. Shears, C.J. Kirk and M. A-Titheradge, 1986, Changes in free

cytosolic calcium and accumulation of inositol phosphates in iso lated heptatocytes

by [leu]enkephalin, Biochem. 1. 238, 537.

Li, P.P., D. Sibony, M.A. Green and J.J. Warsh, 1993, Lithium modulation of

phosphoinositide signaling system in rat cortex: Seledive effect on phorbol ester

binding, J. Neurochem. 6 1, 1722.

Manji, H.K.. R. Etchebemgary, G. Chen and J.L. Olds, 1993, Lithium decreases

membrane-associated protein kinase C in hippocampus: Selectivity for the n

isozyme, J. Neurochem. 6 1,2303.

Mao, J., D.D.Pnce and D.J. Mayer, 1994, Thermal hyperalgesia in association with

the developrnent of morphine tolerance in rats: Role of excitatory arnino acid

receptors and protein kinase C, J. Neurosci. 14, 230 1. Mao, J.R., D.D. Price, L.L. Phillips, I. Lu and D.J. Mayer, 1995, Increased in protein

kinase C gamma immunoreactivity in the spinal cord of rats associated with

tolerance to the analgesic effects of morphine, Brain Res. 677, 257.

Marek, P., S. Ben-Eliyahu, A.L. Vaccarino and J-C. Liebeskind, 199 1a, Delayed

application of MK-80 1 attenuates development of morphine tolerance in rats,

Brain Res. 558, 163.

Marek, P., S. Ben-Eliyahu, M. Gold and J.C. Liebeskind, 199 1b, Excitatory amino

acid antagonists (kynurenic acid and MK-801) attenuate the development of

morphine tolerance in rats, Brain Res. 547, 77.

Marks, F., B. Hanke, O. Thastrup and G. Furstenberger, 1991, Stimulatory effect of

thapsigargin, a non-TPA-type turnor promoter, on arachidonic acid metabolism in

the mutine keratinocyte line HEL30 and on epidermal ce11 proliferation irz vivo as

compared to the effects of phorbol ester TP4 Carcinogenesis 12, 149 1.

Mayer, D.J., J. Mao and D.D. Price, 1995, The development of morphine tolerance

and dependence is associated with translocation protein kinase C, Pain 6 1, 365.

Mayer, M.L. and G.L. Westbrook, 1987, The physiology of excitatory arnino acids in

the vertebrate central nervous system, Prog. Neurobiol. 28, 197.

Monaghan, D.T., R.J. Bridges and C.W.Cotman, 1989, The excitatory amino acid

receptors: Their classes, pharmacoiogy and distinct properties in the fùnction of

the central nervous system, Ann. Rev. Pharmacol. Toxicol. 29, 365. Narita, M., Y. Feng, M. Mahura, b. Hoskins, and I.K. Ho, 1994%A protein kinase

inhibitor, H-7, inhibits the developrnent of tolerance to opioid antinociception, Eur.

J. Pharmacol. 27 1, 543.

Nanta, M., M. Makimura, Y. Feng, B. Hoskins and I.K. Ho, 1994b, Influence of

chronic morphine treatment on protein kinase C activity: cornparison with

butorphanoi and implication for opioid tolerance, Brain Res. 650, 175.

Okajima, F., H. Tomura and Y. Kondo, 1993, Enkephaiin activated the phospholipase c/c:+ systern through cross-talk between opiate receptors an P,-purinergic or

bradykinin receptors in NG 108- 15 cells: A permissive role for pemissis toxin

sensitive G-proteins, Biochem. J. 290, 24 1.

Paxinos, G. and C. Watson, 1986, The rat brain in stereotaxic coordinates (Academic

Press Inc., San Diego).

Pellegrini-Giarnpietro, D.E., M. Ruggiero, S. Giannelli, S. Chiarugi and F. Moroni,

1988, Morphine withdrawal in vitro: potentiation of agonist-dependent

polyphosphoinositide breakdown, Eur. J. Pharmacol. 149, 297.

Percheilet, E.M., H.U. Gali, X.M. Gao and J.P. Perchellet, 1993. Ability of the non-

phorbol ester-type tumor-promoter thapsigargin to rnirnic the stimulatory effects of

12-O-retradecanoylphorbol-13-acetate on ornithine decarboxylase activity,

hydroperoxide production, and macromolecule synthesis in mouse epidermis in

vivo, Intematl. J. of Cancer 55, 1036.

Periyasarny, S. and W. Hoss, 1990, Kappa-opiate receptors stimulate phosphoinositide

turnover in rat brain, Life Sci. 47, 2 19. RafEa, R.B.,C.D. Comelly and R.P. Martinez, 1992, Opioid efficacy is linked to the

Lithium chloride-sensitive inositol- 1,4,5-trisphosphate-restorablepathway, Eur. J.

Pharmacoi. 217, 221.

RafFh, R.B.and R.F.Martinez, 1992, Morphine antinociception is mediated through a

Lithium chloride-sensitive, tP3 restorable pathway, Eur. J. Pharrnacol. 2 1 5, 3 57.

Schoepp, D.D. and P.J. Con. 1993, Metabotropic glutamate receptors in brain

function and pathology, Trends Phamacol. Sci. 14, 13.

Sladeczek, F., J.P. Pin, M. Récasens, J. Bockaert and S. Weiss, 1985, Glutamate

stimulates inositol phosphate formation in striatal neurones, Nature 3 17, 7 17.

Smart, D., G. Smith and D.G. Lambert, 1994, mu-Opioid receptor stimulation of

inositol ( 1.4.5) trisphosphate formation via a pertussis toxin-sensitive G protein, J.

Neurochem. 62, 1009.

Sugiyama, H., 1. Ito and C. Hirono, 1987, A new type of glutamate receptor linked to

inositol phospholipid metabolism, Nature 325, 53 1.

Thastmp, O., P.J.Cullen, B.K.Drobak, M.R. Hanley and A.P. Dawson, 1990,

2- Thapsigargin, a tumor promoter, discharges intracellular Ca stores by specific

inhibition of the endoplasrnic reticulum Ca2(+)-ATPase, Proc. Natl. Acad. Sci.

USA 87,2466.

Trujillo, K. A. and H. Akil, 199 1, Inhibition of morphine tolerance and dependence by

the NMDA receptor antagonist MK-80 1, Science 25 1, 85. Tsu, R.C., J.S.C. Chan and Y.H. Wong, 1995, Regulation of multiple effectors by the

cloned delta-opioid receptor - stimulation of phospholipase C and type II adenylyl

cyclase, J. Neurochem. 64, 2700.

Young, E.A., N.H. NeEand M. Hadjiconstantinou, 1994, Phorbol ester administration

transiently increases aromatic L-amino acid decarboxylase activity of the mouse

striatum and rnidbrain, J. Neurochem. 63, 694.

Yu. V.C. and W. Sadee, 1986, Phosphadidylinositol turnover in neuroblastoma cells:

regulation by bradykinin, acetylcholine, but not mu- and delta-opioid receptors,

Neurosci. Lett. 71, 219.

Zhang, L.J., X. J. Wang and J.S. Han, 1990. Phorbol ester suppression of opioid

analgesia in rats, Life Sci. 47, 1775. Figure 1

A) Mean time spent exhibiting teeth chattering and writhg (tirne in withdrawal) during the 40 minute withdrawal period for morphine-dependent rats treated chronically with either vehicle, chelerythrine or thapsigargin i.c.v (n = 4 to 8 per group). There appears to be a dose-related inhibition of withdrawal symptorns by both chelerythrine and thapsigargin. Planned cornparisons indicated that 10 nmoVday of either chelerythrine (F(1.30) = 8.64, P < 0.0 1) or chapsigargin (F(1,3 1) = 5.32, P <

0.05) significantly decreased the time spent in withdrawal.

* significantly different fiom control, P < 0.05, planned comparison

B) Frequency of counted symptoms dunng the 40 minute withdrawal period for morphine-dependent rats treated chronically with either vehicle, chelerythrine or thapsigargin i.c.v. (n = 4 to 8 per group). Kniskal-Wallis test for non-parametnc data showed no significant effects of either chelerythrine (H(3,23) = 3.57, P > 0.05) or thapsigargin (H(3,23) = 0.52, P > 0.05). chelerythrine thapsigargin vehicie

o. 1 1 10 dose (nmovday)

vehicle chelerythrine thapsigargin i.c.v. treatment Figure 2

Percent of time spent in non-withdrawal (arnbulating, rearing, resting and grooming) and withdrawai behaviours (teeth chattering and writhing combined) for rats treated chronically with either icv. 10% DMSO, or 10 nmoVday of chelerythrine or thapsigargin i. C.V. alone (non-dependent; A) or with s-c. morphine (morphine- dependent; B) during the 10 minutes przor to the injection of naloxone ( I mgkg, s.~.); and in rats treated chronically with i.c.v. vehicle, chelernhrine or thapsigargin either alone (non-dependent; C) or with sac.morphine (morphine-dependent; D) dunng the

IO minutes afrer the injection of naloxone (n = 4 to 8 per group). ANOVA indicated a significant i.c.v. treatment by morphine treatment interaction (F(2,3 1) = 5.08, P <

0.05) and a significant morphine treatment by time interaction (F(2,3 1) = 14.50, P <

0.01) for the percent of time spent resting. There was a significant time block effect for percent of time spent ambulating (F(1.3 1) = 222.80, P < 0.0 1). The percent of time spent grooming also showed a significant effect of time block (F(1,3 1) = 34.79. P

< 0.01). ANOVA indicated a significant effect of morphine treatment (F(1,3 1) = 4.99,

P < 0.05) and time (F(1,3 1) = 53.32, P c 0.0 1) for percent of time spent rearing. For percent of time spent in withdrawal there was a significant i.c.v. treatment by morphine treatment by time interaction (F(2,3 1) = 3.97, P < 0.05). See Results section for a description of specific dserences indicated by Tukey's post-hoc tests. a A non-dependent: pre-naloxone B dependent: pre-naloxone

vehicle chelerythrine thapsigargin vehicle chelerythrine thapsigargin i.c,v. ireaiment i.c.v. treatment

groom mli~ lear

C non-dependent: post-naloxone D dependent: post-naloxone

vehicle chelerythrine thapsigargin vehicle chelerythrine thapsigargin i.c.v. treatment i,c.v. treatment Figure 3

A) Mean time spent exhibiting teeth chattering and writhing (tirne in withdrawal) during the 40 minute withdrawal for morphine-dependent rats given a single i.c.v. injection of either vehicle, or 0.1 or 1 nrnol chelerythrine, or 0.05 or 0.5 nmol thapsigargin 10 minutes pnor to precipitation of withdrawal (n = 4 to 8 per group).

Although 1 nrnol of chelerythrine appeared to decrease the time spent in withdrawal, the results failed to reach sigruficance (planned cornparison F(1.13) = 3.60, P = 0.08).

B) Frequency of counted symptoms dunng the 40 minute withdrawal period for morp hine-dependent rats given a single i. c.v. injection of vehicle, chelerythrine or thapsigargin (n = 4 to 8 per group). Kniskal-Wallis ANOVA for non-parametric data indicated a significant effect of chelerythrine (H(2,16) = 7.3 1, P < 0.05). Post-hoc

Mann-Whitney U-tests indicated that 1 nrnol chelerythnne significantly increased the fiequency of counted symptoms (P < 0.05).

* significantly different from control, P < 0.05, Mann-Whitney U-test fkequency of counted symptome time in withdrawal (sec + sem) $+ Figure 4

Percent of time spent in non-withdrawal (ambulating, rearing, resting and grooming) and withdrawal (teeth chattenng and writhing combined) behaviours for rats given a single i.c.v. injection of vehicle, 1 mol chelerythrine or 0.5 mol thapsigargin alone (non-dependent; A) or with S.C. morphine (dependent; B) during the

10 minutes przor to i.c.v. injection; in rats given a single i.c.v. injection of vehicle, chelerythrine or thapsigargin alone (non-dependent; C) or with S.C. morphine

(dependent; D) dunng the 1O minutes afler i.c.v. injection but prior to naloxone injection; and in non-dependent (E) and morphine-dependent (F) rats given a single i. c.v. injection during the 10 minutes ufter the injection of naloxone (1 mgkg, S. c.).

ANOVA indicated a significant morphine treatment by time interaction for percent of time spent resting (F(2,58) = 3 1.19, P < 0.0 1). For percent of time spent ambulating,

ANOVA indicated a significant i.c.v. effect (F(2,29) = 5.0 1, P < 0.05) and a significant morphine treatment by time interaction (F(2,58) = 13 .Z3,P < 0.0 1). There was a significant time effect for percent of time spent groorning (F(2,SS) = 1 7.39, P < 0.0 I ).

ANOVA indicated a significant morphine treatment by time interaction for percent of time spent rearing (F(2.58) = 4.10, P < 0.05). ANOVA indicated a significant morphine treatment by time interaction for percent of time spent in withdrawal

(F(2,58) = 59.27, P < 0.01). See Results section for a description of specific differences as indicated by Tukey's post-hoc tests. - - percent of time in behaviour percent of time in behaviour percent of tirne in behaviour P bd u

percent of time in behaviour percent of time in behaviour percent of tirne in behaviour ai

Y m "PI CHAPTER 5: Studies of the role of mGluR desensitization in opioid dependence Thus far, we have seen that chronic antagonism of mGluRs and related intracellular second messenger systems concurrently with morphine treatment attenuated the development of dependence. These results suggest that perhaps dunng chronic opioid treatment changes may be elicited in glutamatergic, as well as opioidergic, neurons.

There is evidence that glutarnatergic neurons become desensitized upon repeated activation. In dissociated hippocarnpal neurons, it has been show that high bath concentrations of glutamate and quisqualate produce desensitization in voltage clamp expenments (Fagni, Beaudry and Lynch, 1983; Kiskin, Krishtal and

Tsyndrenko, 1986; Mayer and Westbrook, 1987). AIthough desensitization in t hese expenments was attributed to NMDA and AMPA receptors, mGluRs are also present in the hippocampus (Masu el al, 1994), therefore a component of the desensitization to glutamate or quisqualate may be attributable to desensitization of mGluRs.

Moreover, pre-exposure of cultured cerebellar neurons to glutamate has been shown to decrease PI hydrolysis induced by subsequent application of glutamate (Catania,

Aronica, Sortino, Canonico and Nicoletti, 199 1). indicating that mGluRs in these cells may be desensitized.

It has been shown that while acute administration of p-opioids decreases

CAMP production (Childers, 199 1) and PI hydrolysis (Barg et al, 1994; Barg et al,

1992; Johnson et al, 1994), during chronic administration both CAMP production

(Childers, 1 99 1) and PI hydrolysis (Dixon et al, 1990) retum to near control levels, while dunng withdrawal both CAMP production (Childers, 199 1) and PI hydrolysis (Busquets et al, 1995; Mao et al, 1995; Pelligrini-Giarnpietro et al, 1988) are enhanced. This suggests the eiicitation of cornpensatory mechanisms related to the functioning of these intraceilular second messenger systems. Opioid receptors and mGluRs are sirnilady distributed in several brain regions (Mansour et al, 1995; Masu et al, 1994). suggesting that they may be CO-localizedwithin the same cells, and thus share pools of intracellular second messengers (adenylate cyciase and CAMP, as well as phosphoinositides). Therefore, we speculate that activity at one type of receptor may modulate activity at the other receptors via the shared second messengers. Thus, we hypothesize that chronic activation of opioid receptors may indirectly elicit a desensitization of mGluRs, which in tum may contnbute to the development of opioid dependence.

In this chapter, we investigated the role of mGluR densitization in opioid dependence by exarnining whether the severity of opioid withdrawal can be attenuated by adrninistenng mGluR agonists just pior to the precipitation of withdrawal. We non-selectively activated mGluRs with 1 -arninocyclopentane- 1,)-dicarboxylic acid

((1 S,3R)-ACPD) (Palmer et al, 1989; Schoepp et al, 199 1a; Schoepp et al, 19%;

Schoepp et al, 199 1b; Watkins and Collingridge, 1994). Group 1 mGluRs were selectively activated by (RS)-dihydroxyphenylglycine (DHPG) (Schoepp et al, 1994;

Watkins and Collingridge, 1994). Group iI mGluRs were selectively activated by

( i S,3 S)-ACPD and 2S, l'RJ1K3 R)-2-(2',3'-dicarboxycyclopropy1)ycine @CG-IV)

(Ishida et al, 1993; Pin and Duvoisin, 1995; Watkins and Collingridge, 1994). L-AP4 was used to selectively activate group DI mGluRs (Foster and Fagg, 1984; Nakanishi,

1992; Nakajima eî al, 1993; Okamoto et al, 1994; Watkins and Coliingridge, 1994).

We chronically infused morphine S.C. for seven days. On the seventh day of treatment, we injected either (1 S,3R)-ACPD, DHPG, (1 S,3S)-ACPD, DCG-IV, or L-

AP4 into the laterai ventricle 10 minutes before the precipitation of withdrawal and measured the seventy of abstinence symptorns. We hypothesized that if desensitization of mGluRs was elicited during chronic morphine treatment, at the time of withdrawal, concentrations of endogenous glutamate would not sufficiently activate mGluRs to promote normal functioning, and this may contribute to the occurrence of withdrawal symptoms. However, exogenous administration of agonist may sufficiently activate mGluRs to promote normal functioning, and therefore reduce the seventy of withdrawal symptoms.

As will be discussed, the acute administration of the non-selective agonist

( 1S,3R)-ACPD, as well as the mGluR2/3 selective agonist DCG-IV, significantly decreased the seventy of precipitated withdrawal symptorns. Br. J. Phmucol. (1996) (submitted)

Attenuation of precipitated morphine withdrawal symptoms by acute i.c.v.

administration of a group II mGluR agonist

Marian E. ~undytus'*~and Terence I. ~oderre'*".~

1 Pain Mechanisms Laboratory, Chnical Research Institute of Montreal, 2~epartmentof Psychology, McGill University and 'centre de recherche en sciences neurologiques et départment de médecine, Université de Montréal.

Short litle: DCG-IV attenuates morphine withdrawal

J To whom correspondence should be sent:

Dr. Terence J. Coderre Pain Mechanisms Laboratory Clinical Research Institute of Montreal 1 10 Pine Avenue West Montreal, Quebec, Canada H2W 1R7

Telephone: (5 14) 987-5750 FAX: (5 14) 987-5585 Summary

1. We previously showed that chronic i.c.v. antagonism of mGluRs concurrently with

S.C. morphine significantly attenuated precipitated withdrawal symptoms. Conversely. acute i.c.v. injection of a seiective group II mGluR antagonist just pior to the precipitation of withdrawai exacerbated abstinence symptoms.

2. In the present study, we show that acute i.c.v. administration of the non-selective mGluR agonist (1 S,3R)-ACPD, as well as the group II selective agonist DCG-IV. significantly attenuate the severity of precipitated withdrawal symptoms.

3. From these results we hypothesize that chronic opioid treatment may indirectly induce a desensization of group 11 mGluRs, which contributes to the development of dependence.

Kry<.ordF: morphine; dependence; (1 S,3R)-ACPD; (1 S,3S)-ACPD; DHPG; DCG-IV;

L-AP4; metabo tropic glutamate receptor; precipitated withdrawal Recently, we have shown that chronic i.c.v. antagonism of metabotropic

glutamate receptors (rnGluRs) concurrently with morphine treatment attenuates the

severity of the precipitated withdrawal syndrome (Fundytus & Coderre, 1994;

Fundytus et al, 1996). These results suggest that chronic opioid administration may

elicit changes in glutarnatergic, as well as opioidergic, neurons. Activation of p-opioid

receptors and mGluRs both aect CAMP production and PI hydrolysis. Acute

activation of p-opioid receptors decreases PI hydrolysis (Barg et al, 1992, 1993,

19941, while activation of group 1 mGluRs stimulates PI hydrolysis (Hayashi el al,

1994; Schoepp & Conn, 1993). Activation of p-opioid receptors or group II

(mGluR2 and 3) or group III (mGluR4,6, 7 and 8) mG1uR.s decreases CA.

production (C hilders, 199 1; Hayashi et al, 1994).

Both opioid receptors and mGluRs are directly coupled to these intracellular

second messengers via guanine nucleotide (G)proteins, and opioid receptors and

mGluRs are similarly distributed in the brain (Mansour et al, 1995; Masu et al, 1999,

suggesting that they may be CO-localizedwithin the sarne cells. Thus, opioid receptors

and mGluRs may share cornmon pools of intracellular second messengers, and activity

at one type of receptor may modulate activity at the other receptors via actions on

second messengers. Whereas acute administration of p-opioids decreases CAMP

production and PI hydrolysis, during chronic administration activity in both systems

retums to near control levels (Childers, 199 1; Barg et al, 1992, 1993, 1994; Dixon et ai, 1990), suggesting that compensatory mechanisms are elicited. Given the

sirniiarities of the intraceilular events triggered by opioid and mGluRs, and given that chronic antagonism of mGluRs reduces morphine withdrawal symptoms, we hypothesize that, via actions on second messengers, chronic opioid administration may induce a change in the sensitivity of mGluRs, which contributes to the development of opioid dependence.

In the present study, we examined the possibility that desensitization of mGluRs contributes to morphine dependence by assessing the ability of acute i. c.v. treatment, just prior to the precipitation of withdrawal, with both non-selective and subtype-selective mGluR agonists, to inhibit naloxone-precipitated morphine withdrawal. It is expected that if'mGluR desensitization plays a role in opioid dependence, then acute administration of mGluR agonists just pior to the precipitation of withdrawal should reduce the seventy of abstinence symptoms.

In the present study we compared the effect of 1-arninocyclopentane- 1,3- dicarboxylic acid ((1 S,3R)-ACPD), which is a non-selective mGluR agonist that is ten times more potent as an agonist at mGluRs than ionotropic glutamate receptors

(Palmer et al, 1989; Schoepp et al, 199la, b; Schoepp et al, 1992; Watkins &

Collingridge, 1994) on precipitated withdrawal symptoms in chronic morphine treated rats. The erects of (1 S,3R)-ACPD were then compared to a series of subtype- selective mGIuR agonists. Group 1 mGluRs were selectively activated by (RS)- dihydroxyphenylglycine (DHPG) (Schoepp et al, 1994; Watkins & Collingridge,

1994). Group II mGluRs were selectively activated by (1 S,3 S)-ACPD and

2S, 1'~2'R,3'R)-2-(2',3'-dicarboxycyclopropyI)ycie @CG-IV) (Ishida et al, 1993;

Pin & Duvoisin, 1995; Watkins & ColIingridge, 1994). To activate group DI mGluRs, we used L-AP4 (Nakanishi, 1992; Nakajima et al, 1993; Okamoto et al, 1994;

Watkins & Collingridge, 1994).

In these studies, we found that acute administration of either the non-selective agonist ( l S.3 R)-ACPD, or the mGIuR213 selective agonist DCG-IV, significantly reduces the severity of the precipitated morphine withdrawal syndrome.

Materials and Methods

Sirbjects and Surgery

Subjects were male Long Evans rats (Charles River, PQ), housed 2-4 per cage, maintained on a 12: 12 hour 1ight:dark cycle (lights on at 06:00), with food and water available ad Iibitzim. Rats weighed 280-350 gram at the time of surgery.

On Day O, each rat was anaesthetized with sodium pentobarbital (Sornnotol,

MTC Pharmaceuticals, 60 mg kgm'),and a 23 gauge stainiess steel guide camula was implanted above the lateral ventncle (AP = - 1.3 mm and L = - 1.8 mm from bregma, and V = -3 .O mm from the top of the skull; Paxinos and Watson, 1986) so that the 30 gauge injection camula extended 1 mm below this into the lateral ventricle. While the rat was still under pentobarbital anaesthesia, one unpnmed (i-e. not yet pumping)

AlzetQ osmotic pump containing 50 mg ml" morphine sulfate was implanted subcutaneously (s.c.) on the back. On Day 1, rats were briefly anaesthetized with halothane and a second Alzet@ osmotic pump containing 70 mg ml-' morphine sulfate was implanted S.C. on the back of each rat. This two day pump implantation procedure was used to prevent mortality from a lethal concentration of morphine before any tolerance has developed. Drugs

Morphine sulfate (gift from Sabex, Quebec) was continuously infused

subcutaneously (s.c.) at a rate of 10 pl hf' for a total dose of 36.65 pmol daf '. The

rnGluR agonists (1 S,3R)-ACPD, (1 S,3S)-ACPD, DHPG, L-AP4 and DCG-IV were

al1 obtained fiom Tocris Cookson (Bristol, üK). (1 S,3R)-ACPD (n = 18), (1 S.3 S)-

ACPD (n = 18). DHPG (n = 16) and L-AP4 (n = 17) were given

intracerebroventricu1arIy (i.c.v.) as an acute injection in a volume of 4 pl at a dose of

either O (vehicle) (n = 15 ), 0.12, 0.6 or 3 nrnol, while DCG-IV (n = 1 1) was given

i.c.v. in a dose of either 4.8 or 24 pmol, since higher doses would produce a non-

selective activation of NMDA receptors (Ishida et al, 1993).

Withdrawal Measurement

The severity of abstinence symptoms was assessed on the seventh day of

morphine treatment following an s.c. injection of 1 mg kg-' naloxone. Behaviour was observed for IO minutes before i.c.v. injection of either vehicle or one of the mGluR

agonists, 10 minutes after i.c.v. injection but before naloxone, and for 40 minutes after

the injection of naloxone. Teeth chattering and writhing were timed and combined

into a time spent in withdrawal score. The severity of diarrhea was assessed by the amount of weight lost during the test session. Seventy of eye twitch and salivation

were rated on a 4 point scaie where O = absent and 3 = severe. To assess the effects of mGluR agonists on generai behaviour, time spent in withdrawal and non-withdrawal

(resting, arnbulating, rearing, groorning) behaviours were compared between dependent (given morphine) and non-dependent rats given either i.c.v. vehicle or the highest dose of one of the agonists for the 10 minutes prior to i.c.v. injection, the 10 minutes after i.c.v. but before naloxone injection, and the 10 minutes afler naloxone injection.

S~atis~icul A naiysis

Time spent in withdrawal and weight loss were analyzed using a 1-way

ANOVA perfonned on each mGluR agonist group with dose as the factor. Significant results were further anaiyzed using post-hoc LSD t-tests. Time spent in withdrawal and non-withdrawal behaviours were analyzed with a mixed ANOVA with morphine treatment and i.c.v. treatment as between subjects factors, and time block as a repeated measures factor, again followed by post-hoc LSD t-tests on significant results. Severity of eye twitch and salivation were analyzed with Kniskal-Wallis

ANOVA for non-parametric data, followed by Mann-Whitney U-tests on significant results.

Results

36.65 pmol day' l morphine sulfate produced an intense and reliable withdrawal syndrome as indicated by the presence of abstinence syrnptoms after naioxone injection. As shown in Figure 14the non-selective mGluR agonist (1 S,3R)-

ACPD (F(3.29) = 4.33, P < 0.05) and the mGluRZ3 selective agonist DCG-IV (F(L~)=

26.84, P < 0.01) significantly decreased the arnount of time spent in withdrawal, with

DCG-IV producing the greatest decrease even at the very low doses used. Although

DHPG = 1.67, P > 0.05) and L-AP4 (F(3,zn= 1.92, P > 0.05) appeared to decrease time spent in withdrawal at the highest dose used, the results failed to reach statistical significance. The severity of diarrhea was attenuated in DCG-IV-treated rats, as indicated by a reduction in weight loss (F(=, = 1 1.45, P < 0.0 1) in Figure 1B.

Figure 1C shows the severity of eye twitch, and Figure 1D shows the severity of salivation. DCG-IV decreased the severity of both eye twitch (H«26) = 1 1.03, P C

0.0 1) and salivation fi226) = 10.6 1, P < 0.0 1), while the non-selective agonist

(lS,3R)-ACPD decreased the severity of eye twitch @&333) = 8.69, P < 0.05). The mGluR4 selective agonist L-AP4 significantly increased the severity of eye twitch

(H(j,32)= 1 1.69, P < 0.05).

To assess the effects of acute i.c.v. injection of mGluR agonists on general behaviour, non-withdrawal and withdrawal behaviours were compared in non- dependent and morphine-dependent rats dunng the 10 minutes prior to i.c.v. injection, the 10 minutes alter i.c.v. injection but before naloxone injection, and the 10 minutes following naloxone injection. Generally, rats were more active earlier in the test session, and by the third time block, after the injection of naloxone, non-dependent rats spent most of their time resting. Morphine-dependent rats behaved very similarly to non-dependent rats until after the injection of naloxone, at which time they spent more time in withdrawal (P < 0.05, LSD t-test). There were no effects of i.c.v. treatment except that (1 S,3R)-ACPD and DCG-IV treated rats spent significantly less time in withdrawal than vehicle treated rats (P < 0.05, LSD-t-test; data not shown).

Discussion

The present study shows that the severity of the precipitated morphine withdrawal syndrome was sigruficantly decreased by acute i.c.v. administration of the non-selective mGluR agonist (1S,3R)-ACPD. Although the mGluR1/5 agonist DHPG and the mGluR4 agonist L-AP4 appeared to decrease withdrawal at the highest doses used, the results failed to reach statistical sigruficance. The seledive rnGluRU3 agonist DCG-IV almost completely eliminated teeth chattering and writhing, as well as significantly attenuating the severity of diarrhea (indicated by weight loss). The effects of these agonists on general (non-withdrawal) behaviour were negligible. Although

DCG-IV effectively attenuated the severity of withdrawal, the mGluRU3 agonist

(1 S.3 S)-ACPD failed to do so. This may be due to the fact that (1S,3S)-ACPD is less selective, and stimulates PI hydrolysis and CAMP formation via actions at group 1 mGluRs, as well as decreasing CAMP formation via group II mGluRs (Schoepp and

Conn, 1993).

Although there is some evidence that DCG-IV is an agonist at NMDA receptors at concentrations above 10 p.M NI VZ~TD(Ishida et al, 1993), Our highest dose was only 0.006 W. Therefore, the doses we used were most likely selective for mGluR2/3 receptors. Thus, the efficacy of DCG-IV can be attributed to its agonist action at mGluR2/3. Moreover, we previously found that while chronic i.c.v. administration of a selective rnGluRZ3 antagonist significantly decreased the seventy of precipitated withdrawal, acute i.c.v. administration of this antagonist just prior to the precipitation of withdrawal significantly increased the seventy of abstinence syrnptoms (Fundytus, Ritchie and Coderre, 1996). Taken together, these results support the hypothesis that group II mGluRs may be desensitized during chronic morphine treatment. Chronic antagonism of a receptor is generally believed to induce up-regulation of the receptor. Thus, perhaps chronic antagonism of group II mGluRs induced an up-regulation of these receptors, which counterbalanced the desensitizing effects of morphine treatment, resulting in normally functioning receptors, and thus a reduction of withdrawal symptoms. Also, acute administration of DCG-IV in morphine treated rats may have restored function to group II mGluRs which were no longer sensitive to physiological concentrations of glutamate, enabling these receptors to inhibit CAMP production, and thus decrease the severity of withdrawal.

Our present results provide further support for the hypothesis that mGluR activity may contribute to the developrnent of morphine dependence. Furthermore.

Our results are of clinical interest. Activation of mGluRs may aid in alleviating withdrawal symptoms of patients and thus make detoxification safer and easier.

Acknowledgemen ts

This work was supported by MRC grants #MT- 1 1045 and MT- 13236 and a grant from the Stairs Memorial Fund awarded to T.J.C.M.E.F. was supported by an

FCAR studentship. The authors wish to thank Dr. Y. Ohfune for his generous gifi of

DCG-IV. The authors wish to thank Jerinifer Ritchie for her valuable assistance. References

BARG,J., M.M. BELCHEVAAND C.J.COSCIA (1992). Evidence for the implication of

phosphoinositol signal transduction in mu-opioid inhibition of DNA synthesis. J.

Nerrrochem., 59, 114-1152.

BARG, J., M.M. BELCEVA,R. ZIMLICHMAN,R. LEVY,D. SAYA, R.J. MCHALE, F.E.

JOHNSON, C.J. COSCIAAND 2. VOGEL(1 994). Opioids inhibit endothelh-

2+ mediated DNA synthesis, phosphoinositide turnover, and Ca mobilization in rat

C6 glioma cells. J. Neurosci., 14, 5858-5864.

BARG,J., S.Y. NAH,R. LEVY,D. SAYAANDZ. VOGEL(1993). Modulation of

thymidine incorporation by kappa-opioid ligands in rat spinal cord-dorsal root

ganglion CO-cultures.Bruin Res., 629, 10% 1 14.

C~IILDERS,S.R. ( 199 1. Opioid receptor-coupled second messenger systems. LIfr

Sci., 48, 1 99 1-2003.

DEON, W., Y .-W. TINGAND P.L. CHANG(1990). The effects of morphine on

norepinephrine-stimulated phosphatidylinositol response in rat cerebral cortex.

Prog. Ch.Biol. Res., 328, 279-282.

F~YTUS,M.E. & CODERRE,T.J. (1994). Morphine dependence is affected by

activity at rnetabotropic, as well as ionotropic (NMDA), glutamate receptors. Br.

J. Pharmacol., 113, 1215-1220.

Fw~s,M.E., RITCHIE, J. & CODERRE,T.J. (1996). Attenuation of morphine

withdrawal symptoms by subtype selective metabotropic glutamate receptor

antagonists. Br. J. Pharmacol.., submitted. HAYASHI,Y., SEKIYAMA, N., NAKANISHI,S., JANE, D.E., SUNTER,D.C., BIRSE,E.F.,

UDVARHELYI, P.M. & WATK~NS,J.C. ( 1994). hdysis of agonist and antagonist

activities of p henylglycine derivatives for different cloned metabotropic glutamate

receptor subtypes. J. Neurosci., 14, 3 3 70-3 3 77.

ISHIDA, M.,SAITOH, T., SHIMAMOTO,K., O-, Y. & SHINOZAKI, H. (1993). A

novel metabotropic glutamate receptor agonist: marked depression of

monosynaptic excitation in the newbom rat isolated spinal cord. Br. J. Pharmacoi.,

109, 1169-1 177.

MANSOUR,A., FOX, C.A., AICL,H. & WATSON,S.J. (1995). Opioid-receptor mRNA

expression in the rat CNS: anatomical and fûnctional implications. Trends

Nerrrosci., 18, 22-29.

MASU,M., NAUJMA, Y.,MORIYOSHI, K.,ISHII, T., AKAZAWA, C. & NAKANISHI,S.

( 1 994). Molecular characterization of NMDA and metabotropic glutamate

receptors. Am. AW Acad Sci., 707, 153- 164.

NAKA-, Y., IWAKABE, H., AKAZAWA, C., NAWA,H., SHIGEMOTO,R., M~ZUNO,

N. & NAKANISHI,S. (1993). Molecular characterization of a novel retinal

metabotropic glutamate receptor mGluR6 with a high agonist selectivity for L-2-

amino-4-phosphonobutyrate. J. Biol. Ch.,268, 1 1, 868- 1 1, 873.

NAKANISHI,S. (1 992). Molecular diversity of glutamate receptors and implications

for brain function. Science, 258, 597-603.

OKAMOTO,N., HORI, S. AKAZAWA, C.,HAYAsHI, Y., SHIGEMOTO, R., PVL[ZUNO, N. &

NAKANISHI,S. (1994). Molecular characterization of a new metabotropic glutamate receptor mGluR7 coupled to inhibitory cyclic AMP signal transduction.

J. Biol. Chern., 269, 123 1-1236.

PALMER,E., MONAGHAN,D.T. & COW, C.W. (1989). Trm-ACPD, a selective

agonist of the phosphoinositide-coupled excitatory amino acid receptor. Ezîr. J.

Phcmnacol., 166, 585-587.

PAXINOS,G. & WA-[SON, C. (1986). The rat brain in stereotaxic coordinates.

Acadernic Press Inc., San Diego, California.

Pm,J.-P. & Duvios~,R. (1995). Review: Neurotransmitter receptors 1. The

Metabotropic glutamate receptors: Structure and Functions. Nezwophannacol.,

34, 1-24.

SCHOEPP,D.D. & CONN, P.J. (1993). Metabotropic glutamate receptors in brain

fùnction and pathology . Trenak Phumacol. Sci., 14, 13 -25.

SCHOEPP,D.D., GOLDSWORTHY,J., JOHNSON,B.G., SALHOFF, C.R. & BAKER,S.R.

(1 994). 3.5-Dihydroxyphenylglycine is a highly selective agonist for

phosphoinositide-linked metabotropic glutamate receptors in the rat hippocampus.

J. Nezirochem., 63, 769-772.

SCHOEPP.D.D., JOHNSON,B.G. & MONN,J.A. (1992). Inhibition of cyclic AM.

formation by a selective metabotropic glutamate receptor agonist. J Neurochrrn.,

58, 1184-1 186.

SCHOEPP,D.D., JOHNSON,B.G., SALHOFF,CR., MCDONALD, LW. & JOHNSON,

M. V. ( 1 99 1a). In vitro and in vivo pharmacology of trans- and ci.+(?)- 1-amino- 1 -3-cyclo- pentanedicarboxylic acid: dissociation of metabotropic and ionotropic

excitatory amino acid receptor effects. J. Neurochem., 56, 1 789- 1 796.

SCHOEPP,D.D., JOHNSON, B.G., TRUE,R.A. & MON, J.A. (1 Wlb). Cornparison of

( 1S,3 R)- 1-aminocyclopentane- 1,3-dicarboxylic acid (1S,3R -ACPD)- and lR,3S-

ACPD-stimulated brain phosphoinositide hydrolysis. Eur. J. Phammol., 207,

35 1-353.

WA~S,J.C. & COLLINGRDGE,G.L. (1 994). Phenylglycine denvatives as

antagonists of metabotropic glutamate receptors. Trends Pharmacol. Sci., 15, 3 3 3 -

342. Figure 1

A) Mean time spent in withdrawal (teeth chattering and writhing combined) during

the 40 minute withdrawal period in morphine-dependent rats given an acute i. c.v.

injection of either vehicle (@; n = 15), or 0.12, 0.6 or 3 nmol of ( I S,3R)-ACPD (O:

n = 18), (1 S.3S)-ACPD (O; n = 18), DHPG (A; n = 16) or L-AP4 (V;n = 17). or

4.8 or 24 pmol of DCG-IV (O; n = 1 1).

* significantly less than control (P < 0.05, LSD t-test)

B) Mean weight loss during the 40 minute withdrawal penod in morphine-dependent

rats given an acute i.c.v. injection of either vehicle (e),or 0.12, 0.6 or 3 nmol of

( iS,3R)-ACPD (O), (lS,3S)-ACPD (O), DHPG (A) or L-AP4 (V), or 4.8 or 24

pmol of DCG-IV (O).

* significantly less than control (P < 0.05, LSD t-test) mean weight loss (gis.e.m.) mean time in withdrawal (se- s.e.m.) Figure 2

A) Average severity of eye twitch dunng the 40 minute withdrawal period in

morphine-dependent rats given an acute i.c.v. injection of either vehicle (I),( 1S,

3R)-ACPD (s),(1S73S)-ACPD (e), DHPG p),L-AP4 p)or DCG-IV

+ significantly less than control (P < 0.05, LSD t-test)

# significantly greater than control (P < 0.05, LSD t-test)

B) Average severity of salivation during the 40 minute withdrawal period in

morphine-dependent rats given an acute i.c.v. injection of either vehicle (I),( 1S.

3R)-ACPD (=),( 1S,3S)-ACPD (E),DHPG w),L-AP4 p) or DCG-IV

-

* significantly less than control (P < 0.05, LSD t-test)

CHAPTER 6: Involvement of &opioid recepton in the development of

morphine tolerance and dependence Throughout the previous studies, we have concentrated on the contribution of mGluRs to the development of morphine dependence. We have stressed that mGluRs act via the same intracellular second messenger systems, narnely CAMP production and

PI hydrolysis, as opioid receptors.

Morphine is primariiy a p-opioid agonist, but also has activity at 6- and K- opioid receptors (Takemori and Portoghese, 1987). The production OFCAMPis decreased by activation of p- and 6opioid receptors (Childers, 199 l), as well as by activation of group II (mGluR2 and 3) and group III (mGluR4, 6. 7 and 8) mGluRs

(Hayashi et al, 1994). Thus, morphine's p-opioid effects on CAMPproduction may be sufficient to explain coincidental effects at group II and III mGluRs elicited by chronic morphine treatment.

However, activation of p-opioid receptors decreases PI hydrolysis (Barg et al,

1994; Barg et al, 1992; Johnson et al, 1994), whereas activation of group I (mGluR1 and 5) mG1uRs increases PI hydrolysis (Schoepp and Conn, 1993). Thus, chronic activation of p-opioid receptors would not be expected to elicit compensatory mechanisms in mGluRs positively coupled to PI hydrolysis. But, morphine also activates 6- and K-opioid receptors (Takemori and Portoghese, 1987). It has been show that activation of 6- and K-opioid receptors produces an increase in PI hydrolysis (Barg et al, 1993; Feng et al, 1994; Jin et al, 1994; Leach et al, 1986;

Okajima et al, 1993; Periyasamy and Hoss, 1990; Smart et al, 1994; Tsu et al, 1995). an action simiiar to that produced by activation of group 1 mGiuRs. Thus, it is possible that 6- or K-opioid receptors contribute to opioid dependence through an action on PI hydrolysis. If this is true, then opioid dependence should be attenuated by

selective blockade of 6- or K-opioid recepton, which acts to prevent alterations in PI

hydrolysis. We have already shown that chronic inhibition of the products of PI

hydrolysis concurrently with morphine treatment attenuated the severity of the

precipitated morphine withdrawal syndrome. Moreover, although the role of K-

opioids in the development of opioid tolerance and dependence is unclear, it has been

shown that antagonism of 6-opioid receptors with the relatively selective 6-opioid

antagonists naltrindole, naltrindole-5'-isothiocyanate and naitriben during morphine

treatment attenuates the development of tolerance and dependence (Abdelharnid et al,

199 1; Miyarnoto et al, 1993). However, because naltrindole and naltriben also have

significant antagonistic activity at p- and K-opioid receptors (Portoghese et al, 199 l),

a definitive role for the 6-opioid receptors in the developrnent of morphine tolerance

and dependence has not been established.

In the next study, we attempt to veriQ the role of 6-opioid receptors in the

development of morphine tolerance and dependence by using the highly selective 6-

opioid antagonists TIPP and TPP[v] (Schiller et al, 1992; Schiller et al, 1993).

Because neither TPP nor TIPP[y] show activity at p- or K- opioid receptors (Schiller et al, 1992; Schiller et al, 1993), any effects we obtain on the development of

morphine can be attributed to their actions at Gopioid receptors.

Prior to any treatment, we obtained a morphine dose-response curve in the tail flick test. Then, concurrently with systemic morphine, we infùsed i.c.v. either naltrindole, TIPP or TIPP[w]. Mer six days of treatment, we again obtained a morphine dose-response curve in the tail flick test in rats given one of the antagonists and morphine. We compared the ED50 values fkom the two tests to determine whether tolerance had developed to morphine's analgesic effects. To assess the degree of dependence, we precipitated withdrawal (with S.C. naloxone) on the seventh day of treatment and measured the severity of abstinence symptoms.

In these studies, we have show that while naltrindole, TIPP and TIPP[iv] al1 attenuate the developrnent of dependence, only TIPP[y] prevents the development of tolerance. Attenuation of morphine tolerance and dependence with the highly seleetive 6 opioid receptor antagonist TIPP(yr1

Marian E. Fundytus ", Peter W. Schiller "', Michelle Shapiro *', Grazyna Weltrowska b , Terence J. Coderre

" Pain Mechanisms Laboratory, Clinicai Resed Instztute of Montreal. Monrreal, Quebec, Canada b Chernicd Biology and Peptide Resemch Laborutories, Ciinicai Research Inslitzrtr of Montreai, Montreai, Quebec, Canada Department of Psychology. McGill University, Montreal. eebec, Canaab J Department of Medicine. University of Montreai, Monfieal, Quebec, Canadn Department of Phannacology, University of Montreal, Monireal, Quebec, Canada

* Corresponding author. Pain Mechanisms Laboratory, Clinical Research Institute of Montreal, Montreal, Quebec Canada H2W 1R7. Tel: (5 14) 987-5750; fax: (5 14) 987-5624. Abstract

We examined the effects of i.c.v. treatment with naltrindole, and the two highly selective peptide 8-opioid receptor antagonists H-Tyr-Tic-Phe-Phe-OH (TIPP) and H-

Tyr-Ticy[CH2-NH]-Phe-Phe-OH(TIPP[v]), on the development of morphine tolerance and dependence. Each treatment significantly decreased naloxone- precipitated withdrawal, with TIPP[v] reducing the most symptoms. TIPP[w], but neither naltrindole nor TPP, attenuated the development of analgesic tolerance in the tail-flick test. These results suggest that 5-opioid receptors are criticaily involved in the development of morphine tolerance and dependence.

Keywords: Analgesia; Opiate; 6-opioid receptor antagonist; Opioid withdrawal;

Abstinence 1. Introduction

Opioid dmgs such as morphine are commonly used in the management of pain; however, their usefulness is cornpromised by the developrnent of tolerance and dependence. Morphine is primarily a p-opioid receptor agonist, but it also acts at both

6- and K-opioid receptors (Takemori and Portoghese, 1987). Moreover, it has been proposed that activation of an allosterically coupled 6-opioid receptor modifies activity at the p-opioid receptor (Rothrnan and Westfall, 1982; Tiseo and Yaksh, 1993). Due to this interaction between p- and 6-opioid receptors, it has been hypothesized that activation of 6-opioid receptors may play a sigdïcant role in the development of opioid tolerance and dependence (Abdelhamid et al., 199 1).

Blockade of 6-opioid receptors with the relatively selective non-peptide antagonists naltnndoie, naltrindole-5'-isothiocyanate(5'-NTII) and naltnben has been shown to attenuate the development of morphine tolerance and dependence in rnice

(Abdelhamid et al., 199 1; Miyamoto et al., 1993). However, naltrindole and naltriben also possess significant antagonist potencies at both p- and K-opioid receptors in the guinea pig ileum assay (Portoghese et al., 199 1). The peptide H-Tyr-Tic-Phe-P he-OH

(TIPP) (Schiller et al., 1992) and its pseudopeptide analogue H-Tyr-Ticy[cH~-

NHIPhe-Phe-OH (TIPP[y]) (Schiller et al., 1993) are potent and highly selective 6- opioid receptor antagonists which exhibit no p- or K-opioid receptor antagonistic properties in the guinea ilium assay at concentrations up to 10 W. The purpose of the present investigation was to verify the specific involvement of 6-opioid receptors in the development of morphine tolerance and dependence by comparing the effects of naltnndole and the two highiy selective 6-opioid receptor antagonists, TIPP and

TIPP[\y1, in chronic morphine-treated rats.

2. Materials and methods

2.1. Subjr cts and surgery

Subjects were male Long Evans rats (280-3 50 g). Rats were housed 2-4 per cage, with food and water available ad libitum.

On day 0, rats were anaesthetized with sodium pentobarbital (Somnotol, MTC

Pharmaceuticals), and 23-gauge stainless steel cannulae, attached to model 200 1 Alzet osmotic mini pumps filled with one of the 6-antagonist solutions or saline, were implanted stereotaxically in the laterai ventrical (i.c.v.) (AP = - 1.3 mm and L = - 1.8 mm from bregma, and V = -3.8 mm from the skull surface). During surgery, the first of two model 2ML 1 Alzet pumps containing 60 mg/d morphine sulfate solution was implanted subcutaneously (s.c.) on the back; the second was implanted on day 1, while the rats were briefly anaesthetized with halothane. A second group of rats was given i.c.v. treatment alone (non-dependent) to assess the effects of the 6-opioid receptor antagonists on general behaviour.

2.2. Dngs

TIPP and TIPP[y] were synthesized as descnbed previously (Schiller el al.,

1992, 1993). Naltrindole was obtained fiom Research Biochemicals (Natick, MA,

USA). gopioid receptor antagonists were uifused i.c.v. at a rate of 1 pVh in the following doses: naltrindole (10 nmoVday), TIPP (80 nrnoYday) and TIPP[v] (40 nmol/day). These doses of antagonists were chosen to be equipotent based on K, values obtained in the mouse vas deferens assay (Schiller et al., 1993). Morphine suifate (Sabex, Montreal, PQ, Canada) was continuously delivered S.C. at a rate of 10 pVh for a total dose of 36.65 prnoVday.

2.3. Tolerance measurement

To assess the development of analgesic tolerance, the ED5~of morphine in a radiant heat tail-flick test was determined both pnor to chronic morphine treatment

(pretreatment) and fier 6 days of morphine treatment (chronic). A baseline tail-flick latency was measured for each rat pnor to each morphine trial. For the pretreatment and chronic trials rats were injected S.C.with morphine sulfate (0, 0.5, 1 or mglkg for pretreatment tests; 0, 1, 2, 4 or 8 mgkg for post-treatment tests) and tail-flick latencies were measured for 90 min post-injection at 10 min intervals. A percent maximum possible effect (%MPE) score for each tail-flick latency measurement following the injection of morphine was calculated using the following formula:

%MPE = [(test latency - baseline latency)/cutoff -baseline)]X 100. A cutofF latency of

IO s was used to prevent tissue injury. The area under the curve (AUC) for the

%MPE scores over the 90 min test penod was calculated for each rat. The AUC scores were then used to calculate the EDso values.

2.4. Withdrawai measurement

Precipitated abstinence symptoms were assessed on the seventh day of treatment after injection of naloxone (1 mgkg SC.). For 10 min before and 40 min aAer naloxone injection, withdrawal symptoms were assessed by measuring the amount of time spent teeth chattering and writhing. To ensure that chronic i.c.v. treatment with the gopioid antagonists had no effect on generai behaviour, the tirne spent in non-withdrawal behavours (ambulating, rearing, grooming and resting) was also assessed during the 10 min pnor to the precipitation of naloxone in both morp hine-de pendent (given i. c.v. treatments and S. c. morphine) and non-dependent

(given i-c.v. treatments alone) rats. The average severity of checked signs, eye twitch and salivation, was rated on a 4 point scale, where O = absent and 3 = severe, at the end of each 10 min penod.

3. Results

Chronic treatment with 36.65 pmoVday morphine sulfate produced an intense and reliable withdrawal syndrome, as evidenced by the large amount of time spent teeth chattering and writhing, as well as the occurrence ofjumps. wet dog shakes, eye twitch and salivation in i.c.v. saline-treated rats (Fig. 1). Al1 three 6-opioid receptor antagonists significantly decreased the arnount of time spent in withdrawal (teeth chattering and writhing combined) during the 40 min withdrawal period (Fig. 1A).

Both naltrindole and TIPP[y] significantiy decreased the severity of eye twitch, while

TLPP[w] also significantly decreased the severity of salivation (Fig. 1B).

Fig. 1C illustrates the percent of time spent in non-withdrawal (ambulating, reanng, groorning and resting) as well withdrawal (teeth chattering and writhing) behaviours during the 10 min prior to the precipitation of withdrawal (baseline period) for morphine-dependent rats. Fig. ID shows the percent of time spent in non- withdrawal and withdrawal behaviours for non-dependent rats. As seen in Fig. 1C and

D, there was a significant effect of morphine treatment, with morphine-dependent rats resting and groorning more and ambulating less than non-dependent rats. However, there were no significant differences between i.c.v. treatment groups, indicating that chronic i.c.v. treatment with 6-opioid receptor antagonists had negligible effects on general behaviour either in morphine-dependent (Fig. 1C) or non-dependent (Fig. 1D) rats.

TIPP[y], but neither naitrindole nor TIPP, attenuated the development of analgesic tolerance, with an EDso following chronic morphine treatment (EDsa = 0.36 mgkg) that was unchanged fiom the baseline EDso(EDso = 0.35 mgkg) (Table 1 ).

4. Discussion

The present results suggest that activity at 6-opioid receptors is critical to the development of morphine tolerance and dependence. Al1 three ô-opioid receptor antagonists, naltrindole, TIPP and TIPP[y], effectively decreased the amount of time spent teeth chattering and writhg during the precipitated withdrawai syndrome, while having no effect on general behaviours such as ambulating, rearing, grooming and resting. Both naltrindole and TIPP[y] attenuated the severity of eye twitch while

TIPP[w] also devreased the severity of salivation. However, none of the i.c.v. 6- opioid receptor antagonist treatments afTected the frequency of jurnps or wet dog shakes (data not shown). Jumps and wet dog shakes have been shown to be mediated by brain areas around the fourth ventricle (Laschka et al., 1976). Because we infused very small volumes into the lateral ventricle, the 6-opioid receptor antagonists may have been absorbed by brain areas around the lateral ventricle, and therefore not have been able to diffuse adequately to the fourth ventricle. In TIPP[w]-treated rats, the EDSofor morphine in the td-fick test following 6 days of morphine treatment was unchanged from the pretreatment value. Thus, chronic inhibition of 6-opioid receptors in the brain with TIPP[v] was effective in suppressing the development of both tolerance and dependence during chronic systemic morphine treatment.

Even though we used equipotent doses, TIPP[v] was more effective than either naltrindole or TIPP in attenuating the developrnent of tolerance and dependence.

TIPP[rv] may have been more effective because it is more selective than naltrindole, and more stable in physiological medium than TIPP (Schiller et al., 1993). Unlike previous reports in mice, we did not find an attenuation of analgesic tolerance with naltrindole in rats, suggesting a possible species difference.

It rernains to be determined whether activation of gopioid receptors is critical oniy to the developrnent of morphine tolerance and dependence, or whether this activation is important to the development of opioid tolerance and dependence in general. If this is a phenornenon relating to opioids in general, the development of opioid tolerance and dependence may be attenuated by use of opioid compounds with rnixed p-opioid receptor agonist/&opioid receptor antagonist properties.

Acknowledgements

This work was supported by gants from MRC Canada (MT-11-45) and the

Stairs Memonal Fund to T.J.C.and MRC (UI-12356), NIDA @A-04443) and Astra

Pain control to P.W.S. M.E.F. is supported by a studentship from the FCAR Quebec. References

Abdelharnid, E.E., M. Suitana, P. S. Portoghese and A.E. Takemori, 199 1, Selective

blockage of delta opioid receptors prevents the development of morphine tolerance

and dependence in mice, J. Pharmacol. Exp. Ther. 258,299.

Laschka, E., H. J. Teschernacher, P. Mehraein and A. Herz, 1976, Sites of action of

morphine involved in the development of physicai dependence in rats,

Psychopharmacologia (Berlin) 46, 14 1.

Miyamoto, Y., P.S. Portoghese and A.E. Takemori, 1993, lnvolvernent of delta*

opioid receptors in the development of morphine dependence in mice, J.

Pharmacol. Exp. Ther. 264, 1 14 1.

Portoghese, P.S., H.Nagase, K.E. Maioney Huss, C.-E. Lin and A.E. Takemori,

199 1, Role of spacer and address components in peptidomimetic 6 opioid receptor

antagonists related to naltrindole, J. Med. Chem. 34, 17 15.

Rothrnan, R.B.and T.C. Westfall, 1982, Allosteric couplhg between morphine and

enkephalin receptors in vitro, Mol. Pharmacol. 21, 548.

Schiller, P.W., TM.-D. Nguyen, G. Weltrowska, B.C. Wilkes, B.J. Marsden C.

Lernieux and N.N. C hung, 1992, Differential stereochemicai requirements of p vs.

6 opioid receptors for ligand binding and signal transduction: development of a

class of potent and highly 6-selective peptide antagonists, Proc. Natl. Acad. Sci.

USA 89, 11871. Schiller, P.W., G. Weltrowska, T.M.-D. Nguyen, B.C.Wilkes, N.N. Chung and C.

Lernieux, 1993, TIPP[w]: a highly potent and stable pseudopeptide 6 opioid

receptor antagonist with extraordinary 6 selectivity, J. Med., Chem. 36. 3 182.

Takernori, A.E. and P. S. Portoghese, 1987, Evidence for the interaction of morphine

with kappa and delta opioid receptors to induce analgesia in p-funaltrexamine-

treated mice, J. Pharmacol. Exp. Ther. 243, 91.

Tiseo. P.J. and T.L.Yaksh, 1993, Dose-dependent antagonism of spinal opioid

receptor agonists by naloxone and naltrindole: additional evidence for 6-opioid

receptor subtypes in the rat, Eur. J. Pharmacol. 236, 89. TABLE 1: Morphine EDw for rats phor to morphine treatment, and for rats in each i.c.v. treatment group foilowing 6 days of morphine treatment.

Treatment EDm (mgkg) (95% CL)

Pretreatment 0.35 (0.33 - 0.53)

Vehicle, post-treatment 0.8 1 (0.30 - 1.66)'

Naltnndole, post-treatment 0.98 (0.06 - 3.16)"

TIPP, post-treatment 1.25 (0.22 - 3.47)"

TPP[v], post-treatment 0.36 (0.05 - 0.81)

" significantly greater than pretreatment (l?< 0.05, Student's t-test) Figure 1

A. Mean time spent teeth chattering and writhing dunng the 40 min withdrawal penod for rats treated with 36.65 pmoVday morphine sulfate (s.c.) and either vehicle, naltrindole, TIPP or TIPP[v] (i.c.v.). ANOVA reveaied a sigruficant main effect of i.c.v. treatment for time spent in withdrawai (teeth chattering and writhing) (F(3,35) =

6.63, P < 0.01). * indicates a significant difference From the vehicle control group (P

< 0.05,LSD t-test).

B. Average severity of checked signs (eye twitch and salivation, maximum = 3) for rats in each treatment group. Kruskal-Wallis ANOVA for non-parametric data indicated a significant effect of i.c.v. treatment for both eye twitch (H(3,39) = 137, P

< 0.O 1) and salivation (H(3,39) = 9.92,P < 0.05). * indicates a significant difference frorn the vehicle control group < 0.05,Mann-Whitney U-test).

C. Percent of time spent in non-withdrawal (ambulating, rearïng, grooming and resting) and withdrawal (teeth chattering and writhing combined) behaviours during the 10 min pnor to precipitation of withdrawal for morphine-dependent rats in each i. C.V. treatment group.

D. Percent of tirne spent in non-withdrawal and withdrawal behaviours dunng the 10 min pnor to the precipitation of withdrawal for non-dependent rats. ANOVA of the data in C and D indicated that there were no differences in any of the behaviours between different i.c.v. treatment groups. ANOVA indicated a significant effect of morphine treatment on resting (F(1,54) = 7.11, P < 0.0l), ambulating (F(1,54) = 25 -54. P c 0.01) and groorning (F(1,54) = 4.58, P < 0.05). Morphine-dependent rats tendend to rest and groom more, while ambularing less, than non-dependent rats. % of time epent in behaviour

Oh of time epent in behaviour Average eeverity of checked eigne CHAPTICR 7: General Discussion 1. Summary of results in relation to previous data

Recently, several groups of investigators have shown that chronic administration of NMDA receptor antagonists concurrently with morphine treatment attenuates the development of tolerance and dependence (Marek ei ai, 199 1 a; Marek et al, 1 99 1b; Trujillo and Akil, 199 1), suggesting the involvement of the endogenous excitatory amino acid (EAA) glutamate. Because glutamate also acts at two other ionotropic receptors, AMPA and kainate, as well as at a group of rnetabotropic receptors (mGluRs) (Mayer and Westbrook, 1987a; Monaghan et al, 1989), our first goal was to determine if these other receptors also play a role in the development of morphine tolerance and dependence. We were particularly interested in mGluRs because, similarly to opioid receptors, they are directly coupled, through G proteins, to

CAMP production and PI hydrolysis.

In Chapter 2, we established a role for mGluRs, as well as NMDA receptors, in the development of morphine dependence. Chronic i.c.v. administration of the rnGluR antagonists L-AP3 and (S)-4CPG, as well as the NMDA antagonist MK-80 1, concurrently with systemic morphine treatment significantly attenuated the development of morphine dependence, as indicated by a decreased severity of precipitated withdrawal symptoms, with negligible effects on non-withdrawal behaviours. In contrast, there was no effect of an AMPAkainate antagonist, GYKI

52466. Furthemore, acute i.c.v. administration of EAA antagonists just prior to the precipitation of withdrawal had no effect on the severity of abstinence symptoms

(unpublished observations). This is in contrast to other investigators, who have observed a reduction in abstinence symptoms with single injections of NMDA antagonists just pnor to the precipitation of withdrawal (Koyuncuoglu et al, 1977;

Koyuncuoglu et al, 1992; Rasmussen et al, 199 1a; Rasmussen et al, 199 1b; Tanganelli et al, 199 1 ). The discrepancy may be due to route of antagonist administration (brain versus systemic), or type of morphine administration (chronic infusion versus repeated injections). Although there was a trend for chronic i.c.v. antagonists to attenuate the development of tolerance (data not shown), our results failed to reach statistical significance. This is also in contrast to other investigaton who found that repeated injections of EAA antagonists significantly decreased the development of tolerance

(Marek et al, 199 1a; Marek et al, 199 1b; Trujillo and Akil, 199 1). Although we saw some effect on tolerance, Our failure to reach statistical significance may have been due to subject variability, or to route of antagonist administration (brain versus systemic).

Perhaps by giving very srna11 amounts only in the brain, we did not produce enough of a change to affect the spinaily-mediated opioid effects and development of tolerance at that site.

In Our first study we used L-AP3 and (S)-4CPG as rnGluR antagonists. L-

AP3, although the most comrnonly used mGluR antagonist, is non-selective and also shows some activity at NMDA receptors (Birse et al, 1993). (S)-4CPG is a phenylglycine derivative which selectively antagonizes group 1 mGluRs, which are positively coupled to PI hydrolysis, but has a secondary action whereby it is an agonist at group II mGluRs, which are negatively coupled to CAMPproduction (Eaton et al,

1993; Hayashi et al, 1994; Watkins and Collingridge. 1994). Thus, it is not entirely clear whether the chronic antagonism of group 1 mGluRs (and inhibition of PI hydrolysis), or the chronic activation of group II mGluRs (and inhibition of CAMP) was primarily responsible for the efficacy of (S)-4CPG in this experiment.

Thus, in Chapter 3, we examined the role of group II and III mGluRs, which are negatively coupled to CAMPproduction (Hayashi et al, 1994). as are opioid receptors (Childers, 199 1). in the development of morphine dependence. Chronic i.c.v. infusion of the non-selective mGluR antagonist MCPG (Hayashi et al. 1994) concurrently with morphine treatment significantly attenuated the development of morphine dependence. Moreover, selective chronic antagonism of group U mGluRs with MCCG (Jane et al, 1994), as well as selective antagonism of group III mGluRs with MAP4 (Jane et al, 1994), also significanlty attenuated the development of morphine dependence. suggesting a role for mGluRs negatively coupled to CAMP production in the development of opioid dependence. In contrast, acute i.c.v. administration of these mGluR antagonists just prior to the precipitation of withdrawal failed to decrease the severity of abstinence syrnptoms. Rather, acute MCCG treatment significantly increased the tirne spent in withdrawal. These results also suggest that the group U agonistic effects of chronically adrninistered (S)-4CPG were not likely to be responsible for its effective attenuation of morphine dependence.

In Chapter 4, we examined whether changes in PI hydrolysis may contribute to the development of morphine dependence. Group 1 mGluRs are positively coupled to

PI hydrolysis, and we previously showed that chronic antagonism of group 1 rnGluRs with (S)-4CPG significantly attenuated the development of morphine dependence. Although other investigators reported that administration of H-7 or GM 1 ganglioside attenuates the developrnent of opioid tolerance and dependence (Maldonado et al,

1995; Mayer et d,1995; Narita et al, 1994a), neither H-7 nor GM 1 gangiioside are selective, and both affect several intraceiiular messengers such as PKC, PLAz, PKA, cm-dependent protein kinase, and activity at voltage-gated caZ+channels (Hidaka et al, 1984; Bressler et al, 1994; Carlson et al, 1994; Hungund et al, 1994). Thus, it is not clear whether inhibition of PKC is the primary effect of these agents which decreases morphine tolerance and dependence, and therefore we assessed the effects of a highly selective inhibitor of PKC on morphine withdrawal. We showed that selective chronic inhibition of PKC with chelerythrine (Herbert et al, 1990) significantly attenuated the severity of precipitated withdrawal syrnptoms.

Furthemore, we dso showed that chronic inhibition of intracellular ca2' with thapsigargin (Thastrup et al, 1990) significantly attenuated the developrnent of morphine dependence. The efficacy of inhibiting products of PI hydrolysis at attenuating morphine dependence supports the hypothesis that compensatory mechanisms may be elicited in the PI systern dunng chronic opioid treatment.

Morphine is primarily a p-opioid agonist, and activation of p-opioid receptors results in a decrease in PI hydrolysis (Barg et al, 1994; Barg et al, 1993 ; Barg et al, 1 992;

Johnson et al, 1994). Moreover, dunng chronic administration of opioids PI hydrolysis is similar to untreated controls (Dixon et al, 1990), or somewhat increased over control levels (Mao et al, 1995; Mayer et al, 1999, and dunng withdrawal PI hydrolysis is greatly enhanced (Busquets et ai, 1995; Mao et al, 1995; Pellegrini- Giampietro et al, 1988). Given these findings and our own results, we hypothesize that during morphine treatment, compensatory mechanisms are elicited to increase PI hydrolysis to near control levels. Chronic antagonisrn of group 1 mGluRs would keep the level of PI hydrolysis at a low level, and thereby may counteract the homeostatic mechanisms working to increase PI hydrolysis that are elicited by chronic morphine treatment. Because overall PI hydrolysis would be maintained at a low level, the increase in PI hydrolysis seen during withdrawal would be dampened, and abstinence symptoms would be less severe.

In Chapter 5, we exarnined the effeas of mGluR agonists on morphine withdrawal. Because mGluRs and opioid receptors both act via the adenylate cyclase/cAMP and PI intracellular second messenger systems, and mGluRs and opioid receptors are distnbuted in sirnilar brain regions (Mansour et al, 1995; Masu et al,

1994) (suggesting that they may be CO-localizedwithin the sarne cells), we hypothesized that chronic opioid treatment may elicit changes in glutamatergic, as well as opioidergic, receptor systems. Repeated administration of glutamate has been shown to produce desensitization of both ionotropic EAA receptors (Fagni et al,

1983 ; Kiskin et al, 1986; Mayer and Westbrook, 1987) and mGluRs (Catania et al,

199 1 ) in neuronal ceIl cultures. Acute activation of group II and III mGluRs (Hayashi et al. 1994), as well as opioid receptors (Childers, 199 l), leads to inhibition of CAMP production, but CAMP production retums to near control levels dunng chronic administration of opioids (Childers, 1991). This compensation in CAMP production indicates that receptors which are negatively coupled to CAMP production may be desensitized so that their activity is decreased, and CAMP inhibition is reduced dunng chronic opioid treatment. Then du~gwithdrawal, when opioid inhibition of CAMP is removed, CAMP production is enhanced, suggesting that activity at receptors like group iI and ILI mGluRs, which are negatively coupled to CAMPproduction, may not be sufficient to maintain the decrease in CAMP. This suggests that a desensitization of group II or 111 mGluRs may contribute to opioid withdrawai syrnptoms by allowing increases in CAMPproduction. Acute i.c.v. injection of the group II selective agonist

DCG-IV just pnor to the precipitation of withdrawal signincantly attenuated, and in fact virtually elirninated, abstinence symptoms, supporting this hypothesis. Thus, exogenous administration of agonist may produce the necessary level of activation to reduce CAMP production and alleviate withdrawal syrnptoms. Although the non- selective agonist (1 S,3R)-ACPD also sigruficantly reduced the severity of withdrawal, neither the group 1 selective agonist DHPG nor the group III selective agonist L-AP4 significantly affected the withdrawal syndrome. Therefore, the effects of (1 S,3R)-

ACPD may also be explained by its effects at group II mGluRs.

In Chapter 6, we examined the involvernent of 8opioid receptors, which are positively coupled to PI hydrolysis (Barg et al, 1993; Feng et al, 1994; Jin et al, 1994;

Leach et al, 1986; Okajirna et al, 1993; Periyasamy and Hoss, 1990; Smart et al, 1994;

Tsu et al, 1999, as are group 1 mGluRs (Hayashi et al, 1994; Schoepp and COM,

1993), in the development of morphine tolerance and dependence. Moreover, fi- opioid receptors and group 1 mGluRs are distnbuted in some of the same brain areas

(Mansour et al, 1995; Masu et al, 1994), suggesting that they may be CO-localized within the sarne cells and interact via comrnon pools of intracellular second messengers. Therefore, activity at 6-opioid receptors may be eliciting compensatory changes in PI hydrolysis dunng chronic morphine treatment. Although previous studies showed that relatively non-selective 6-opioid antagonists attenuated the development of morphine tolerance and dependence (Abdeihamid et ai, 1 99 1 ;

Miyamoto et ai, 1993). we assessed the specific involvement of 6-opioid receptors by using the highly selective 8opioid receptor antagonists TIPP and TIPP[v], which show no activity at p- or K-opioid receptors (Schiller et ai, 1992; Schiller et al, 1993).

Chronic i.c.v. infusion of both TIPP and TIPP[y] attenuated the development of dependence. while TTPP[w] also prevented the developrnent of tolerance to chronically infùsed morphine. These results support the role of &opioid receptors in the development of morphine tolerance and dependence.

[I. Previous models of EAA receptor contribution and CAMPto morphine dependence

A. NMDA receptor supersensiîiviîy

Based on studies showing that aspartate antagonized some morphine effects

(Koyuncuoglu et al, 1974) and attenuated the developrnent of morphine tolerance and dependence (Koyuncuoglu et a/, 1977), and the fact that morphine decreased L- asparaginase activity (Koyuncuoglu et al, 1979; Koyuncuoglu er ai, 1986).

Koyuncuoglu and colleagues began to develop an EAA related mode1 of opioid tolerance and dependence. They concentrated on NMDA receptors and showed that

NMD A antagonists given just pnor to the precipitation of withdrawal reduced the severity of abstinence symptoms (Koyuncuoglu et al, 1990), while pre-treating rats wit h NMD A antagonists prior to any morphine treatment exacerbated withdrawal symptoms (Koyuncuoglu and Aricioglu, 1991). They hypothesized that opioid bound to EAA receptors (in particular NMDA) and had an antagonistic effect, eliciting EM receptor supersensitivity during chronic opioid treatment (Koyuncuoglu et al, 1990;

Koyuncuoglu and Aricioglu, 199 1). When the opioid action was temiinated by injection of an opioid antagonist, the supersensitive NMDA receptors would then become hyperactive and elicit withdrawal symptorns. Although NMD A receptors may be afFected dunng chronic opioid treatment, there is no evidence to support the contention that opioids bind to NMDA receptors (Jacquet and Squires, 1988), therefore it is unlikely that opioids would have a direct antagonistic action on NMDA receptors.

B. NMDA-induced PKC activiîy

Recently, Mao, Mayer and Pnce (1 995) have designed a mode1 of opioid tolerance in the spinal cord based on an interaction between NMDA receptors and opioid receptors via PKC. They assume that opioid receptors and NMDA receptors are CO-localizedwithin the sarne cells. First, they noted that some investigators observe an increase in PI hydrolysis and PKC activation upon administration of low doses of p-opioid agonists (Smart et al, 1994, 1995; Jin et d,1992). Furthermore,

Mayer et al (1 995) showed that inhibition of the translocation of PKC with GM 1 ganglioside attenuated the development of morphine tolerance and dependence.

Combined with the observations that NMDA antagonists also reduce the development of morphine tolerance and dependence (Trujillo and Akil, 1991), they hypothesized an interaction between opioid receptors and NMDA receptors via PKC activity. They hypothesized that when agonists bind to the p-opioid receptor, PKC is activated via a

G-protein coupled mechanism. The activated PKC could then lead to phosphorylation of NMDA receptor channets, removing the M~**block, and allowing activation of the

NMDA receptor with lower concentrations of glutamate (Mao et al, 1995; Smart and

Lambert, 19%). The opening of the NMDA receptor channel would then permit the influx of ca2', which would fùrther enhance PKC activity, alter gene expression, or elicit ~a~'/calmodulin-dependentnitric oxide (NO) production. NO could then activate ot her protein kinases via cyclic guanosine monop hosp hate (cGMP), or difise out of the ceIl and elicit more glutamate release pre-synaptically. All of these actions of PKC would result in the ceIl being in an excited state. They also specuiate that increased activation of PKC may have a direct effect on the p-opioid receptor, and may uncouple the G-protein fiom the receptor, reducing the effect of the agonist.

Narita el al (1 994) also hypothesize that PKC contributes to the development of opioid tolerance by phosphorylating opioid receptors. There is evidence that PKC potentiates desensitization of both p- and 8-opioid receptors (Mestek, Hurley, Bye.

Campbell, Chen, Tian, Liu, Schulman and Yu, 1995; Ueda, Miyamae, Hayashi,

Watababe, Fukushirna, Sasaki, Iwarnura and Misu, 1995). Smart and Lambert (1996) propose that it is the PKC stimulating effect of p-opioids that is primarily responsible for the development of tolerance, as do Mao et al (1995). This mode1 works very well if one assumes that the concentration of p-opioid agonist is fairly low since it has been shown that low doses of p-opioid agonists stimulate PI hydrolysis (Smart and

Lambert, 1996). However, chronic treatments with opioids are usuaiiy aven in high doses, and other investigators have found that activation of p-opioid receptors with high doses of agonist inhibits PI hydrolysis and subsequent PKC activation (Bug et al,

1992; Johnson et al, 1994; Smart and Lambert, 1996). It should be noted that this model is based on data obtained with morphine administration, and morphine binds to

6- and K-opioid receptors, as well as p-opioid receptors (Takemori and Ponoghese,

1987). Activation of 6- and K-opioid receptors enhances PI hydrolysis and PKC activation (Jin et al, 1992; Srnart et al, 1994). Moreover, the existence of a Ci/& opioid receptor complex has been hypothesized (Rothman and Westfall, 1982).

Perhaps the PI hydrolysis stimulating effects of activation of 6-opioid receptors would fit in with this model. As p-opioid receptors were phosphorylated and desensitized, the PI hydrolysis stimulating effects of 8-opioid receptor activation would become dominant. The enhanced PI hydrolysis would then result in greater concentrations of

PKC, and hence further phosphorylation and desensitize of p-opioid receptors.

C CAMPcompensatory mechanisms

The most common model of opioid tolerance and dependence is based on the interaction of opioid receptors with the adenylate cyclasefcAMP second messenger system. As discussed previously, acute administration of opioid agonists generally leads to an inhibition of CAMPproduction , while dunng chronic opioid treatment production of CAMP retums to near control levels, and during withdrawal CAMP production is enhanced (Childers, 1991), suggesting the elicitation of compensatory or homeostatic mechanisms in this system during chronic opioid treatment. It would be

reasonable to assume that an up-regdation of CAMP during chronic opioid treatment

would lead to an up-regulation of CAMP-dependent protein kinase (PEU), and this

phenomenon has indeed been observed (Nestler and Taliman, 1988; Shiekhattar and

Aston-Jones, 1993). Moreover, it has been shown that administration of H-7

attenuates the development of opioid tolerance and dependence (Maldonado et al,

1995; Nanta et al, 1994a). H-7 is a serindthreonine kinase inhibitor which inhibits

CAMP-dependent protein kinase (PKA), as weli as PKC (Hidaka et al, 1984). Based

on these data, Maldonado, Vaverde, Garbay and Roques (1995) propose that not only

are cornpensatory mechanisms elicited in CAMP production, but this up-regulation

then leads to an up-regulation of Pm which contributes to the development of

morphine tolerance and dependence.

DI. A new model for an EAA receptor mediation of morphine dependence:

critical role of mGluRs

We propose that the cornpensatory or homeostatic mechanisrns elicited in PI

hydrolysis or CAMP production during chronic opioid treatment may be modulated via

mGluRs and opioid receptors. Because brain distributions of mGluRs and opioid

receptors are similar (Mansour et al, 1995; Masu et al, 1994), we suggest that they may be CO-localizedwithin the same cells and share cornmon pools of these intracellular second messengers. Thus, activity at one type of receptor could modulate activity at the other receptors via actions on second messengers. (See Figure 1 for model). We hypothesize that initially high treatment doses with p-opioid agonists decreases PI hydrolysis (and therefore activation of PKC and reiease of intracellular ca2-), while chronic treatment elicits compensatory mechanisms which increase PI hydrolysis to near control levels. This may result because of an increase in the relative act ivity of receptors which are positzveiy coupled to PI hydrolysis, such as group 1 mGluRs and 8-opioid receptors (which have also been proposed to exist in a 0- opioid receptor cornplex), and thus up-regulate PI hydrolysis. Increased activation of

PKC would facilitate phosphorylation of Gi (Childers, 199 l), to which p-opioid receptors are coupled and thus lead to desensitization of p-opioid receptors. The PI stimulating effects of group 1 mGluRs and 8-opioid receptors would then become more dominant. Increased activation of PKC would also rnodulate activity of NMDA receptors, relieving these receptors of the M~~'block and thus potentiate NMDA- mediated activity (Mao et al, 1995; Bleakman et al, 1992; Anikstejn et al, 1992;

Harvey and Collingndge, 1993; Kelso et al, 1992; Chen and Huang, 1992). Up- regulated PI hydrolysis would also lead to increased release of intracellular ca2' (via stimulation by P3),which could be important for several reasons. First, increased concentrations of ca2'i would facilitate activation of ~a~'/calmodulin-dependent protein kinases, leading to further phosphorylation of p-opioid receptors, and thus accelerating desensitization of p-opioid receptors (Maldonado et al, 1995; Mestek el al, 1995). Moreover, increased ~a~'~concentrations may facilitate influx of ca2' through ca2-channels (Lambert, 1993; Lambert, Wojcikiewicz, Safiany, Whitharn, and Nahorski, 1992; Rasmussen, 1986; Smart and Lambert, 1W6), thereby inhibiting the anaigesic effect of the opioid treatment. Our hdings that chronic antagonism of group 1 mGluRs and 6-opioid receptors, as well as chronic inhibition of PKC activation and intracellular caZ+release, reduce morphine withdrawal symptoms support the hypothesis of compensatory changes in PI hydrolysis being mediated by these receptors. We propose that chronic antagonisrn of group 1 mGluRs or 6-opioid receptors would maintain the level of PI hydrolysis at a low level, reducing the amount of activated PKC and release of and thereby reducing desensitization of p-opioid receptors induced by PKC and ~a~+/calmodulin-dependentprotein kinases.

Opioid administration also affects CAMP production. It is generally believed that whereas acute administration of opioids inhibits CAMP production, during chronic administration CAMP production retums to near control levels, and during withdrawal

CAMP production is greatly enhanced (Childers, 199 l), suggesting the elicitation of compensatory mechanisms in this system. We hypothesize that compensation in

CAMP production may in part be induced by desensitization of other receptors that are negatively coupled to CAMP production, such as group II and UI mGluRs. If desensitization of group II and III mGluRs occurs, physiological concentrations of glutamate would not be able to sufficiently activate these mGluRs to provide adequate inhibition of CAMP production during withdrawal to ease symptorns. Chronic antagonism of group II and III mGluRs may induce up-regulation of these receptors.

Thus, proposed desensitization induced by opioid-mediated effects and the up- regulation induced by group II or III rnGluR antagonism may counterbalance, resulting in the normal functioning of group II and III mGluRs. Because group II and III mGluR activity would be maintained at a normal level, physiologicd concentrations of glutamate would sufficiently activate these mG1uRs to provide adequate inhibition of

CAMPproduction. The resultant effect would be an alleviation of abstinence symptoms dunng opioid withdrawal. If mGluR desensitization is not prevented, exogenous application of a group II or 111 agonist pnor to withdrawal rnay be useful in providing the necessary inhibition of CAMP. Our findings that chronic antagonism of group II and HI mGluRs with MCCG or MAP4, as well as acute activation of group II mGluRs with DCG-IV, reduce the severity of morphine withdrawal support this hypothesis.

We can go a step fùrther and speculate that the separate parts of Our mode1 rnay be related. Intracellular second messengers do not act in isolation from one another. First, up-regulated PI hydrolysis rnay affect CAMP production via an iP3- mediated increase in ca2-iconcentration. Increased ca2+;promotes the influx of ~a" through L-type ca2' charnels (Lambert, 1993; Lambert et ai, 1992; Rasmussen, 1986;

Smart and Lambert, 1996), which rnay in tum lead to activation of adenylate cyclase and increased production of CAMP (Hirst and Lambert, 1995; Rasmussen, 1986;

Smart and Lambert, 1996). Phosphorylation of p-opioid receptors and group II and

111 mGluRs rnay also be enhanced by ~a*+/calrnodulin-dependentprotein kinases, leading to the desensitization of these receptors (which are negatively to CAMP production), thus reducing inhibition of CAMPproduction and leading to greater concentrations of CAMP. Up-regulated PI hydrolysis rnay also enhance receptor phosphorylation via a DAG-mediated activation of PKC. Our mode1 draws upon and extends previously proposed models. We agree that PI hydrolysis and CAMP production are intimately involved in the development of opioid tolerance and dependence. The major difference between Our mode1 and others is the suggestion that compensatory or homeostatic mechanisms elicited in PI hydrolysis and CAMP production during chronic opioid treatment may be modulated via mGluRs (as well as 8opioid receptors). Moreover, while other investigators have proposed that it is the PKC stimulating effects of low doses of p-opioids that are important for the development of tolerance, we believe believe this is unlikely since opioid treatrnent doses are usuaiiy quite high. Rather, we propose that PI hydrolysis is more likely up-regulated by the activation of group I mGluRs, and possibly 6-opioid receptors. Moreover, changes in CAMP production rnay also in part be modulated by interactions between opioid receptors and group II and III mGluRs.

IV. Future directions

Our data strongly suggest that mGl& play a significant role in the development of opioid tolerance and dependence. However, the pharmacological tools available to date only differentiate groups of rnGluRs which share sirnilar transduction mechanisms. There are as yet no pharmacological agents which distinguish between the specific receptor subtypes within a group of rnG1uRs. Thus, in future, in would be interesting to employ molecular biologic techniques, with which we could differentiate between the receptor subtypes. Specificdly, we could design antisense treatments which would knock down only one receptor, and assess the effects of this receptor knockdown on the development of opioid tolerance and dependence.

Another interesting question is whether mGluRs contnbute to the development of tolerance and dependence to opioid subtype-selective agonists. In this senes of experiments we made the rats dependent on morphine, which acts at p. 6- and K- opioid receptors. The expenments could be repeated with subtype selective opioid agonists to assess the relative interaction between mGluRs and each of the opioid receptors.

It would aiso be valuable to assess the efficacy of mGIuR antagonists adrninistered spinally. Would brain and spinal mG1uRs mediated dif5erent withdrawal symptoms? Is one route of administration more efficacious than the other? By the sarne token, it would be interesting to evduate the efficacy of systemically administered mGluR antagonists.

Our data also raise some interesting clinical questions. First, cm a safe and effective mGluR antagonist be developed which would make opioid treatment of chronic pain patients more effective? Second, can a safe and effective mGluR agonist be developed which would enable physicians to alleviate withdrawal symptoms in addicted patients whom they were attempting to detoxify?

Thus, this senes of experiments has not only answered some interesting questions, but it has elicited several new and exciting questions to fùrther investigate. Figure 1

This diagram illustrates the receptor function and intracellular changes which

we propose are elicited by chronic morphine treatment Chronic treatment induces a delta-opioid receptor-rnediated (as weli as group 1 mGluR-mediated) stimulation of phosphatidyiinositol (PI) hydrolysis, and subsequently, protein-kinase C (PKC)- rnediated effects (phosphorylation of p-opioid receptors and N-rnethyl-D-aspartate

(NMDA) receptor-associated ion channels (increasing the influx of ~a")). Increased

PI hydrolysis also increases an inositol- 1.45-trisphosphate (IP+mediated release of

~a"from stores on the endoplasmic reticulum (ER), Merincreasing concentrations of intracellular ~a".There rnay be a desensitization of receptors negatively coupled to cyclic adenosine-3'.St-monophosphate(CAMP) production (e.g . group II and III mGluRs). There may also be enhanced influx of ~a"via voltage-gated ~a"channels.

Increased concentrations of intracellular ca2+rnay enhance the activity of both adenylate cyclase and ~a"/calmodullindependent protein kinases. See discussion for a detailed description of the proposed model.

normal activity

+ + + + + + + + + increased stimulation

----a increased inhibition

------decreased inhibition

References ABDELHAMID,EX., SULTANA,M., PORTOGHESE, P.S. AND TAKEMORI,A.E. ( 199 1 ).

Selective blockage of delta opioid receptors prevents the development of morphine

tolerance and dependence in rnice. J Phannacoi. Eip. mer., 258, 299-303.

ABOU-SAMRA,A.B., HARWOOD, J.P., CATT,K.J. & AGLIILERA,G. (1987).

Mechanisms of action of CW and other regulators of ACTH release in pituitary

corticotrophs. Ann. NY Acad Sci., 5 12, 67-84.

ADAMS, J.U. AND HOLTZMAN, S.G. (1990). Tolerance and dependence after

continuous morphine infusion fiom osmotic pumps measured b y operant

responding in rats. Psychophamacolcgy, 100,45 1-458.

AGHAJANIAN,G.K. (1978). Tolerance of locus coemleus neurones to morphine and

suppression of withdrawal response by clonidine. Nature, 267, 186- 189.

AKAOM, H. & ASTON-JONES, G. (1 99 1). Opiate withdrawal-induced hyperactivity of

locus coemleus neurons is substantially mediated by augmented excitatory

amino acid input. J. Neurosci., 11, 3830-3839.

AMBROS~NI,A. & MELDOLESI,J. (1989). Muscarinic and quisqualate receptor-

induced phosphoinositide hydrolysis in primas, cultures of striatal and

hippocampal neurons. Evidence for differential mechanisms of activation. J.

Neztrochern., 53, 825-833.

WSTEJN,L.,OTAN, S. & BEN-AN,Y. (1992). Quisquaiate metabotropic receptors

modulate NMDA currents and facilitate induction of long-term potentiation

through protein kinase C. Etrr. J. Neurosci., 4, 500-505. BARG,J., BELCHEVA,M.M. & COSCIA,C.J. (1992). Evidence for the implication of

phosphoinositol signal transduction in mu-opioid inhibition of DNA synthesis. J.

Nezrrochem., 59, 1145-1 152.

BARG,J., BELCHEV~M.M., ZIMLICHMAN,R., LEVY,R., SAYA,D., MCHALE, R.J.,

JOHNSON,F.E., COSCIA,C.J. & VOGEL,2. (1 994). Opioids inhibit endothelin-

mediated DNA synthesis, phosphoinositide turnover, and ~a2+mobilization in rat

C6 glioma cells. J. Neurosci.. 14, 5858-5864.

BARG, J., NAH, S.Y.,LEVY, R., SAYA, D. & VOGEL,2. (1993). Modulation of

thymidine incorporation by kappa-opioid ligands in rat spinal cord-dorsal root

ganglion CO-cultures.Brain Res., 629, 109- 114.

BARNARD,E.A. & WEY,J.M. (1990). The non-NMDA receptors: types, protein

structure and molecular biology. In i7he Pharmacology of Excitatory Amino

Aci& a TiPS Special Report, Eds. Lodge, D. and Collingndge, G. (Elsevier,

Amsterdam), 82-89.

BEN-ELIYAHU,S., MAREK, P., VACCAEUNO,A. L., MOGIL,J. S ., STERNBERG,W .F. &

LIEBESKIND,J-C. (1 992). The NMDA receptor antagonist MK-80 1 prevents long-

lasting non-associative morphine tolerance in the rat. Brain Res., 575, 304-3 08.

BERL, S., LAJTHA, A. & WAELSCH,H. (1961). Amino acid and protein metabolism.

VI. Cerebral compartments of glutamic acid metabolism. J. Neurochern., 7, 186-

197.

BERRIDGE, M.J. & IRVINE, R.F. (1989). hositol phosphates and ce11 signalling.

Nature, 341, 197-205. BE~ER,B., BOULTER,J., HERMANs-BORGMEYER,I., OSTHEA-GREENFIELD,A.,

DENERIS,E.S., ET.U (1990). Cloning of a novel glutamate receptor subunit,

GluR5-expression in the nervous system during development. Neuron, 5, 5 83 -

595.

BETTLER,B., EGEBJERG,J., SHARMA,G., PECHT,G., HERMANS-BORGMEYER, I.,

MOLL,C., STEVENS,C.F. & HEINEMANN,S. (1992). Cloning of a putative

glutamate receptor: a low affnïty kainate-binding subunit. Neuron, 8, 257-265.

BHARGAVA,H.N., GULATI,A. AND RAMARAO, P. (1989a). Effect of chronic

administration of U-50,488H on tolerance to its pharmacological actions and on

multiple opioid receptors in rat brain regions and spinal cord. J. Phannacol. Eup.

Ther., 251, 21-26.

BFLARGAVA,H.N., RAMARAO, P. AND GULATI,A. (1989b). Effects of morphine in

rats treated chronically with U-50,488K a kappa opioid receptor agonist. Eirr. J.

Pharmacol., 162, 257-264.

BHOOLA,K.D. & PAY, S. (1 986). Opioid inhibition of adenylate cyclase in the

stnatum and vas deferens of the rat. Br. J Pharmacol., 89, 109- 1 18.

BIRSE, E.F., EATON,S.A., JANE, D.E., JONES, P.L.,PORTER, R.H., PoOK, P.C.,

SUNTER,D.C., UDVARHELM,F.M., WHARTON,B., ROBERTS, P.J., SALT, T.E. &

WA-~K~NS.J.C. (1 993). Phenylglycine denvatives as new pharmacological tools

for investigating the role of metabotropic glutamate receptors in the central

nervous system. Neuroscience, 52,48 1-488. BLASIG,J., HERz, A., REINHOLD, K. & ZIEGLGANSBERGER,S. (1 973). Development

of physical dependence on morphine in respect to the and dosage quantification

of the precipitated withdrawal syndrome in rats. Psychophannacoi., 33, 19-3 8.

BLEAKMAN,D., RUSIN, KI., CHARD,P.S., GLAUMS.R. & MILLER,R.J. (1992).

Metabotropic glutamate receptors potentiate ionotropic glutamate responses in the

rat dorsal horn. Mol. Phmacol., 42, 192- 196.

BRADBURY,A.F., SMYTH,D.G. & SNELL,C.R. (1976). Biosynthetic ongin and

receptor conformation of methionine enkephalin. Nafure,260, 165-166.

BRESSLER,J. P., BELLONI-OLIVI,L. & FORMAN,S. (1994). Effect of ganglioside GM 1

of arachidonic acid release in bovine aortic endothelial cells. Life Sci., 54, 49-60.

BROWNSTEIN,M. J. (1 993). A brief history of opiates, opioid peptides, and opioid

receptors. Proc. Nail. Acad Sci. USA, 90, 5391-5393.

Buc~~rr,W.R. (1 964). A new test for morphine-like physical dependence (addiction

liability) in rats. Psychophamacologia, 6, 440-446.

BUSQUETS,X., Escmq P.V.,SASTRE, M.& GARCIA-SEVILLA,J.A. (1 995). Loss of

protein kinase C-alpha beta in brain of heroin addicts and morphine-dependent

rats. J. Neurochern., 64, 247-252.

CALCAGNE~I,D.J. AND HOLTZMAN,S.G. (1991). Delta opioid antagonist,

naltrindole, selectively blocks analgesia induced by DPDPE but not DAGO or

morphine. Pharmacol. Biochem. & Behav., 38, 1 8 5- 190. CARLSON,R.O., MASCO, D., BROOKER,G. & SPIEGEL,S. (1994). Endogenous

ganglioside GM 1 modulates L-type calcium channel activity in N 18 neuroblastorna

cells. J. Neurosci.,14, 2272-228 1.

CARTER,B.D. & MEDZ~HRADSKY,F. (1993). Receptor mechanisms of opioid

tolerance in SH-SYSY human neural celIs. Mol. Phannacol., 43,465-473.

CATANIA, M.V., AROMCA, E., M.A. SORTINO,CANONICO, P.L. & NICOLElTI, F.

( 199 1). Desensitization of metabotropic glutamate receptors in neuronal cultures.

J. Nt?zrrochem., 56, 132% 13 3 5.

CHANG, K.-J. AND CUATRECASAS,P. (1979). Multiple opiate receptors. &O/.

Ch.,254, 26 10-26 18.

CHEN,L. & HUANG, L.-Y.M. (1 992). Protein kinase C reduces M~~*block of

NMDA-receptor channels as a mechanism of modulation. Nature, 356, 52 1-533.

CHEN,L., & HUANG,L.-Y.M. (1991). Sustained potentiation of NMDA receptor-

mediated glutamate responses through activation of protein kinase C by a p opioid.

Nrtrrorr, 7, 3 1 9-3 26.

CHEN,Y., MESTEK,A., Lm, J., HURLEY,J.A. & YU, L. (1993). Molecular cloning

and functional expression of a p-opioid receptor Rom rat brain. Moi. Pharmacol.,

44, 8-12.

C~DERS.S .R. ( 199 1). Opioid receptor-coupled second messenger systems. Life

Sci., 48, i 99 1-2003. CLOWET,D.H., GOLD,G.I. & IWATSUBO,K. (1975). Effects of narcotic analgesic

dmgs on the cyclic adenosine-3',St-monophosphate-adenylatecyclase system in rat

brain. Br. J Pharmacol., 54, 541-548.

COCHIN,J. (1973). Factors iduencing tolerance to and dependence on narcotic

analgesics. In S. Fisher and A.M. Freedman (eds), Opiate Aakiiction: Origins utxi

Treatment, Washington: Winston.

COLLER,H.O.J. & FRANCIS, D.L.(1975 j. Morphine abstinence is associated with

increased brain cyclic AMP. Nature, 255, 159- 162.

COLLIER,H.O. J. (1980). Cellular site of opiate dependence. Nature, 283, 625-629.

COLLER,H.O. J.(1983). Physiological basis of opiate dependence. Dmg h Alcohol

Dep., 111, 15-21.

COLLIER,H.O.J., CUTHBERT,N.J. & FRANCIS,D.L. (198 1). Character and rneaning

of quasi-morphine withdrawal phenomena elicited by methylxanthines. Fed. Proc.,

40, 1513-1518.

COLLINGRIDGE,G.L. & SINGER,W. (1990). Excitatory amino acid receptors and

synaptic plasticity. In The Phannacology of Excitatory Amino Acidî. a TiPS

Special Report, Lodge, D. and Coliingridge, G., Eds., Elsevier: Amsterdam, 42-

48.

COLWELL,C. S. & LEVINE,M.S. (1994). Metabotropic glutamate receptors modulate

N-methyl-D-aspartate receptor function in neostriatai neurons. Neurosci., 6 1, 497-

507. COTMAN,C.W., MONAGHAN,D.T., & GANONG,AH. (1 988). Excitatory amino

acid neurotransmission: NMDA receptors and Hebb-type synaptic plasticity. Am.

Rev. Nezrrosci., 11, 61-80.

COURTNEY,M.J. & NICHOLLS,D.G. (1992). Enteractions between phospholipase c-

coupled and N-rnethyl-D-aspartate receptors in cultured cerebellar granule cells:

Protein kinase C mediated inhibition of N-methyl-D-aspartate responses. Ji

Nezrrochern., 59, 983-992.

COWAN,A., ZHU, X.Z., MOSBERG,H.I., O~~NPLAS,J.R. ANDPoRREcA, F. (1988).

Direct dependence studies in rats with agents selective for different types of opioid

receptor. J. Pharmacol. Exp. mer., 246, 950-955.

CM, SM., CM, B. & PETERSON,E.R. (1986). Cyclic AMP or forskolin rapidly

attenuates the depressant effects of opioids on sensory-evoked dorsal-horn

responses in mouse spinal cord-ganglion explants. Brain Rer ,370, 6 1-72.

CULL-CANDY,S.G. & OGDEN,D.C. (1985). Ion channels activated by L-glutamate

and GABA in cultured cerebellar neurons of the rat. Proc. Royal Soc. of Lorrdon-

Series B: Biol. Sci., 224, 3 67-373.

DAVIES.J. & WATKINS,J.C. (1982). Actions of D- and L-forms of 2-amino-5-

phosphonovalerate and 2-amino-4-phosphonobutyrate in the cat spinal cord.

Brain Res., 235, 3 78-3 86. DESMEULES,J.A., KAYSER, V., GACEL,G., GUILBARD,G. AND ROQUES,B.P. (1993).

The highly selective 6 agonist BUBU induces an analgesic effect in normal and

arthritic rat and this action is not afFected by repeated administration of low doses

of morphine. Brain Res., 611, 243-248.

DIXON,W., TING.Y.-W. & CHANG,P.L. (1990). The effects of morphine on

norepinephrine-stimulated phosphatidylinositoi response in rat cerebral cortex.

Prog. Clin. Biol. Res., 328, 279-282.

DONEVAN,S.D. & ROGAWSKI,M.A. (1993). GYKI 52466, a 2,3-benzodiazepine, is

a highly selective, non-cornpetitive antagonist of AMPAkainate receptor

responses. Neuron, 10, 5 1- 59.

Duvors~,R.M., ZHANG,C. & WONELL.K. (1995). Glutamate receptor expressed

in the retina and olfactory bulb. J. Neurosci., 15, 3075-3083.

EATON,S.A., JANE, D.E., JONES,P.L.ST.J., PORTER,R.H.P., POOK, P.C.-K.,SINER,

D.C.,UDV~IELYI, P.M., ROBERTS,P.J., SALT,T.E. & WATKINS,J.C. (1993).

Cornpetitive antagonism at metabotropic glutamate receptors by (5')-4-

carboxyphenylglycine and (RS)-a-methyl-4-carboxyphenylglycine. Eur. J.

Pharmacol. Mol. Phannacol., 244, 195- 197.

EGEBJERG,J., BETIZER,B., HERMANS-BORGMEYER,1. & HEINEMANN,S. (1 99 1).

Cloning of a cDNA for a glutamate receptor subunit activated by kainate but not

AMPA. Nafzrre, 351, 745-748. ELLIOIT,K., M~NAMI,N., KOLESNIKOV,Y.A., PASTERNAK,G.W. & INTURRISI, C.E.

(1 994). The NMDA receptor antagonists LY2746 14 and MK-80 1, and the nitnc

oxide synthase inhibitor, NG-nitro-L-arginine, attenuate analgesic tolerance to the

mu-opioid morphine but not to kappa opioids. Pain, 56, 69-75.

ENNIS, M., ASTON-JONES,G. & SHIEKHA~AR,R. (1992). Activation of locus

coeruleus neurons by nucleus paragigantocellularis or noxious sensory stimulation

is mediated by intracoerulear excitatory amino acid neurotransmission. Brain Res.,

598, 185- 195.

Emsso~,P.S., HANSSON, E. & RONNBACK,L. (1992). Delta and kappa opiate

receptors in primary astroglial cultures. Part II: Receptor sets in cultures from

various brain regions and interactions with beta-receptor activated cyclic AM..

Neztrochern. Res, 17, 545-55 1.

ERSPAMER,V., MELCHIORRI,P., FALCONERI-ERSPAMER,G.,NEGRI, L., CORSI, R.,

SEVERINI,C., BARRA, D., SWCO, M.& KRIEL, G. (1989). Deltorphins: a

family of naturally occumng peptides with higb affinity and selectivity for delta

opioid binding sites. Proc. Nd.Acad. Sei. USA, 86, 5 188-5 192.

EVANS,C., ~ITH,D., MORRISON, H., ~~AGENDZO,K. & EDWARDS,R. (1992).

Cloning of a delta opioid receptor by hnctional expression. Science, 258, 1952-

1955. the effects of a senes of -phosphonic -carboxylic amino acids on electrically

evoked and amino acid induced responses in isolated spinal cord preparations. Br.

J. Phamtacol., 75, 65-75.

FAGNI, L., BALJDRY,M. & LYNCH,G. (1983). Classification and properties of

excitatory amino acid receptors in hippocampus. 1. Electrophysiological studies

of an apparent desensitization and interactions with drugs which block

transmission. J. Neurosci., 3, 1 538- 1546.

FENG,Y.Z., NARITA,M., MAUMURA,M., HOSKINS,B. & HO, I.K. ( 1994).

Enhancernent of the rate of GTP hydrolysis ifi rat brain by repeated kappa-opioid

treatrnent, Brain Res., 667, 133- 137.

FONNUM,F. (1984). Glutamate: A neurotransmitter in rnamrnalian brain. J.

Nezrrochern., 42, 1- 1 1.

FOSTER,A.C. & FAGG, G.E. (1984). Acidic amino acid binding sites in mammalian

neuronal membranes: their characteristics and relationship to synaptic receptors.

Brazn Res. Rev., 7, 103-164.

FUKAGAWA,Y ., KATZ, J.L. AND SUZUKI,T. (1989). Effects of a selective kappa-

opioid agonist, U-50,488H, on morphine dependence in rats. Eur. J. Pharmacol.,

170, 47-5 1.

FUKUDA, K., DATO,S., MORI, K.,NISHI, M. & TAKESHLM~H. (1993). Primaiy

structures and expression fiom cDNAs of rat opioid receptor delta- and mu-s

ubtypes. FEBS Lett, 327, 3 11-3 14. FUNDWS, M.E. & CODERRE,T.J. (1994). Morphine dependence is affected by

activity at metabotropic, as well as ionotropic (NMDA), glutamate receptors. Br.

J. Phmacol., 113, 12 15- 1220.

FWYTUS, M.E. ( 1992). Central nervous system and peripheral signs of opioid

abstinence. MSc Thesis, McGN University.

FWYT~S, M.E., RTCHIE,J. & CODERRE,T.J. ( 1996). Attenuation of morphine

withdrawal symptoms by subtype selective rnetabotropic glutamate receptor

antagonists. Br. J Phonnacol., submitted.

GASIC,G.P. & HOLLMANN,M. (1992). Molecular neurobiology of glutamate

receptors. Am. Rev. Physioi., 54,507-536.

GMEREK,D.E., DYKSTRA, L.A. AND WOODS,J.H. (1 987). Kappa opioids in rhesus

monkeys. III. Dependence associated with chronic administration. J. Pharmacol.

Exp. Ther.. 242, 4281136.

~LDSTEIN,A., FISCHI, W., LOWNEY, L.I., HUNKAPULER,M. AND HooD, L. (198 1).

Porcine pituitary dporphin: complete amino acid sequence of the biologically

active heptadecapeptide. Proc. Natl. Acad Sci (USA), 78, 72 19-7223.

GOODMAN,R.R. AND PASTERNAK,G.W. (1985). Visuaiization of mu, opiate

receptors in rat brain using a cornputerized autoradiographic subtraction technique.

Proc. Natl. Acad Sci. (USA), 82, 6667-6671. GUTTART,X., THOMPSON,M.A., M&ANTE, C.K.,GREENBERG, M.E. & NESTLER,E.J.

( 1992). Regulation of CAMP response elernent-binding protein (CREB)

phosphorylation by acute and chronic morphine in the rat locus coeruleus. J.

Neurochern., 58, 1 168- 1 17 1.

HALBACH,H. & EDDY,N.B. (1963). Tests for addiction (chronic intoxication) of

morphine type. Bull. W7d Hllh. Org, 26, 139- 165.

HARVEY.J. & COLLINGRIDGE,G.L. (1993). Signal transduction pathways involved in

the acute potentiation of NMDA responses by 1S,3R-ACPD in rat hippocampal

slices. Br. J. Pharmacol., 109, 1085-1090.

HAYASFU,Y., SEKJYAMA, N., NAKANISHX,S., JANE, D.E., SUNTER,D.C., BIRSE,E.F.,

UDVARHELYI, P.M. & WATKINS,J.C. (1994). Analysis of agonist and antagonkt

activities of phenylglycine derivatives for difEerent cloned metabotropic glutamate

receptor subtypes. J. Neuroscz., 14, 3370-3377.

WUM,E., CHANG, K.J., CUATRECASAS,P. AND PASTERNAK,G. W. (1 98 1).

Naloxazone irreversibly inhibits the high affinity binding of [125~]~-Ala2-~-

eu^ ]enkephalin. Life Sci., 28, 2973-2979.

HEJNA,M.H., HOGENBOOM,F., PORTOGHESE,P.S., MULDER,A.H. &

SCHOFFELMEER,A.N. (1989). Mu- and delta-opioid receptor-mediated inhibition

of adenylate cyclase activity stimulated by released endogenous dopamine in rat

neostnatal slices; demonstration of potent delta-agonist activity of bremazocine.

J. Pharmacol. m. ner., 249, 864-868. HERBERT,J.M., AUGEREAU,J.M., GLEYE,J. & MAFFRAND, J.P.(1 990).

Chelerythrine is a potent and specüic inhibitor of protein kinase C. Biochem.

Biophys. Res. Commun., 172, 993-999.

~YMAN,J.S., WILLIAMS, CL., BURKS,TI., MOSBERG, H.I. AND PORRECA,F.

(1 988). Dissociation of opioid antinociception and central gastrointestinal

propulsion in the mouse: Studies with naloxonazine. J. Pharmacol. Erp. Thrr..

245, 23 8-243.

HIDAKA,H., INAGAKI, M., KAWAMOTO,S. & SASAKI Y. (1984). Isoquinoline-

suifonamides, novel and potent inhibitors of cyclic nucleotide dependent protein

kinase and protein kinase C. Biochem., 23, 5036-5041.

HLLER,J.M., PEARSON, J. & SIMON,E.J. (1973). Distribution of stereospecific

binding of the potent narcotic analgesic etorphine in the human brain:

predorninance in the limbic system. Res. Commtrn. Chern. Puthol. & Pharmacol.,

6, 1052- 1062.

HIRST,R.A. & LAMBERT,D.G. (1995). Adenyl cyclase in SH-SY5Y human

neuroblastoma ceiis is regulated by intra- and extra-cellular calcium. Biochem.

Pharmacol., 49, 1633- 1640.

HO, I.K., LOH,H.H. & WAY,E.L. (1972). Effect of cyclic AMP on morphine

analgesia, tolerance and physical dependence. Nature, 238, 3 97-3 98.

HOLLMANN,M., O'SHEA-GREENFIELD,A., ROGERS,S.W. & HEINEMANN, S. (1989).

Cloning by functional expression of a member of the glutamate receptor farnily.

Mure, 342, 643 -648. HONG.O., YOUNG, G.A. AND KHAZAN, N. (1986). Bidirectional cross-tolerance

between methadone (mu)- and ethylketocyclazocine (kappa)-tolerant rats. Dnrg di

Alcohol Dep., 18, 87-96.

HONORE, T., DAVIES,SN., DRWER, J., FLETCHER,E.J., JACOBSEN, P., LODGE,D. &

NIELSEN,F.E. (1988). Quinoxalinediones: Potent cornpetitive non-NMDA

glutamate receptor antagonists. Science, 241, 70 1-703.

HORAN,P.J., WILD, K.D., M~SICKA,A., LIPKOWSKI,A., WSETH,R.C., HRUBY,

V.J., WEBER,S.J., DAVIS,T.P., YAMAMURq H.1. AND PORRECA,F. ( 1993).

Agonist and antagonist profiles of [DMa2, ~lu~]deltor~hinand its [cys4]- and

[sefi]-substituted derivatives: Further evidence of opioid delta receptor

multiplicity. J. Pharrnacof. Exp. Ther., 265, 896-902.

HUGHES,J. (1 975). Isolation of an endogenous compound from the brain with

pharmacological properties similar to morphine. Brain Res., 88, 295-308.

HUGHES.J., SMITH, T.W., KOSTERLITZ,H.W., FOTHERGILL,L.A., MORGAN,B.A.

AM) MORRIS,H.R. (1 975). Identification of two related pentapeptides from the

brain with potent opiate agonist activity. Nature, 258, 5 77-5 79.

HUNGUND,B.L., ZHENG,A., LIN, L. & BARKAI,A.I. (1 994). Ganglioside GM 1

reduces ethanol induced phospholipase A2 activity in synaptosomal preparations

fiom mice. Neurochem. IhternufI., 25, 321-325. ISHIDA, M., SAITOYT.,SHIMAMOTO, K., OHRR~IE,Y. & SHINOZAKI,H. (1993). A

novel rnetabotropic glutamate receptor agonist: marked depression of

monosynaptic excitation in the newbom rat isolated spinal cord. Br. J. Pharmacol.,

109, 1169-1 177.

ISHI1, I., MoRIYOSHI, K., SUGMAR~H., SAKURADA,K., -TANI, H.,YOKOI, M.,

AKAZAWA, C., SKIGEMOTO,R., MIZUNO,N., MASU,M. & NAKANISHI,S. ( 1993).

Molecular characterization of the family of the N-methyl-D-aspartate receptor

subunits. J. Biol. Chem., 268, 2836-2843.

JAFFE, J.H. & MARTIN,W.R. (1990). Opioid analgesics and antagonists. In

Goodman and Gilman 's ''nePharmacological Basis of Therapeuiics ", ed.

Alfred Goodman et al, 8" ed. New York: Pergamon Press.

JANE, D.E., JONES,P.L.ST. J., POO& P.C.-K.,TSE, H.-W. AND WA~S,J-C. ( 1994).

Actions of two new antagonists showing selectivity for dflerent sub-types of

metabotropic glutamate receptor in the neonatal rat spinal cord. Br. J.

Phannacol., 112, 809-8 16.

JACQUET,Y.F. & SQUIRES,R.F. (1988). Excitatory arnino acids: role in morphine

excitation in rat periaqueductal gray. Behav. Brain Res., 31, 85-88.

JAW, S.P., HOSKINS, B. AMI HO, I.K. (1993a). Involvement of 6-opioid receptors in

physicai dependence on butorphanol. Eur. J. Pharmacol. , 240, 67-72.

JAW, S.P., hhKhWM, M., HOSKINS, B. AND HO, 1.K. (1993b). Effects of nor-

binaltorphirnine on butorphanol dependence. Eur. J. PhacoL, 239, 1330 140. JIN, W. LEE, N.M., LOY H. & THAYER, S.A. (1992). Dual excitatory and inhibitory

effects of opioids on intracellular calcium in neuroblastorna x glioma hybrid

NG108- 15 cells. Mol. Pharmacol., 42, 1083- 1089.

JN, W., LEE, N.M.,LOH, H.H. & THAYER,S.A. (1994). Opioids mobilize calcium

from inositol 1,4,5-trisphosphate-sensitive stores in NG108- 1 5 cells. J. Nezrrosci.,

14, 1920- 1929.

JOHNSON, P.S., WANG, J.B., WANG, W.F. & UHL, G.R. (1994). Expressed mu opiate

receptor couples to adenylate cyclase and phosphatidylinositol turnover.

NeuroReporr, 5, 507-509.

JONAS, P. & SAKMANN, B. (1992). Glutamate receptor charnels in isolated patches

from CA 1 and CA3 pyramidal cells of rat hippocampal slices. J. Physiol., 455,

143-171.

KAKIDANI, H.,FURUTANI, Y., TAKAHAsHI, H.,NODA, M., MORIMOTO, Y., ~OSE,

T., ASAI,M., NAYAMA,S., NAKANASHI, S. & NUMA, S. (1982). Cloning and sequence analysis of cDNA for porcine beta-neo-endorphiddynorphin precursor.

Nature, 298, 245-249.

KALSO,E.A., SULLIVAN,A.F., MCQUAY,H.J., DICKENSON,A.H. AND ROQ~S,B.P.

(1 993). Cross-tolerance between mu opioid and alpha-2 adrenergic receptors, but

not between mu and delm opioid receptors in the spinal cord of the rat. J.

Pharmacol. m. mer., 265, 55 1-558. KAPCALA, L.P.,WENG, C.-F. & JUANG, H.-H. (1992). Protein kinase C activators

stimulate beta-endorphin secretion fkom hypothaiamic cells. Bruin Res. Bull., 29,

553-557.

KAYMAKCALAN,S. & WOODS, L.A. (1 956). Nalorphine-induced abstinence

syndrome in morphine-tolerant albino rats. J. Phmacol. Erp. Ther., 117, 1 12-

116.

KENANEN, K.,WISDEN, W., SOMMER, B., WERNER, P., HERB, A., VRDOORN,T.A.,

SAKMANN, B.& SEEBURG, P.H. (1990). A family of AMPA-seledive glutamate

receptors. Science, 249, 556-560.

KELSO, SR.,NELSON, T.E. & LEONARD,J.P. (1992). Protein kinase C-mediated

enhancement of NMDA currents by metabotropic glutamate receptors in Xenupus

oocytes. J. Physiol., 449, 705-5 1 8.

KEFFER, B.L., BEFORT,K., GAVERIAUX-RUE, C. & H-IRTH,C.G. (1992). The 6-

opioid receptor: isolation of a cDNA by expression cloning and pharmacological

characterization. Proc. Natl. Acad Sci. USA, 89, 12048- 12052.

K~NG,A.E., LOPEZ-GARCIA,J.A. & CUMBERBATCH,M. (1992). Antagonism of

synaptic potentials in ventral hom neurones by 6-cyano-7-nitroquinoxaline-2,3-

dione: a study in the rat spinal cord in vitro. Br. J. Phmacoi., 107, 3 75-38 1.

KISKIN, NI., KRISHTAL, O. A. & TSYNDRENKO,A.Y. (1986). Excitatory amino acid

receptors in hippocarnpai neurons: Kainate fails to desensitize them. Newosci.

Lett., 63, 225-230 . KOERNER,J.F. & COTMAN,C. W. ( 198 1). Micromolar L-2-amino-4-

phosphonobutyric acid selectively inhibits perforant path synapses from lateral

entorhinai cortex. Brain Ra., 2 16, 1 92- 198.

KOOB. G.F.. MALDONAW,R. & STINUS.L. (1992). Neural substrates of opiate

withdrawal. TretzA Neurosci., 15, 186-191.

KOSTERLITZ,H. W., LORD, I.A.H. & WATT,A. J. (1972). Morphine receptor in the

myentenc plexus of the guinea-pig ileum. In H.W. Kosterlitz, H.O.J. Collier and

J.E . Vil1 areal (eds) . Agonist and Antagonisî Actions of Narcotic Analgesic Drugs,

MacMallan, London, 45-6 1.

Kov~cs,G.L., NYOLCZAS,N., KRIVAN, M. AND GULYA,K. (1988). Anaigesic and

tolerance-inducing effects of the highly selective delta opioid agonist CD-Pen2, D-

PenSIenekphaiin in mice. Eur. J. Pharmacol., 150, 347-3 53.

KOYUNCUOGLU, H. & ARIGIOGLU, F. (1991). Previous chronic blockade of NMDA

receptors intensifies morphine dependence in rats. Phannacol. Biochem. B

Behav., 39, 575-579.

Komcuo~~u,H., DIZDAR, Y., ARIGIOGLU, F. & SAYIN, U. (1992). Effects of MK-

80 1 on morphine physical dependence: Attenuation and intensification.

Pharmacol. Biochem. & Behav., 43,487-490.

K~YUNCUOGLU,H., G~GOR, M., ENGINAR,N., HATIPOGLU, 1. & HLZAL, A. (1986).

Brain asparaginase, ACE activity and plasma cortisol level in morphine dependent

rats: Effect of aspartic acid and naloxone. Phmacoi. Biochem. & Behav., 25,

953-957. KOYUNCUOGLU, H., GONGOR, M., SAGDUYU, H. & ARIcIOGLU, F. (1990).

Suppression by ketamine and dextromethorphan of precipitated abstinence

syndrome in rats. Phmacoi. Bzochem. & Behav., 35, 829-832.

KOYUNCUOGLU, H.,GONGOR, M.,SAGDUYU, H. & EROGLU, L. (1974). The

antagonistic effects of aspartic acid on some effects of morphine on rats. Eur. J.

Pharmacol., 27, 148- 150.

KOYUN~UOGLU. H.,GÇRJGOR, M.,SAGDUW, H. & EROGLU,L. (1977).

Antagoniring effect of aspartic acid on the development of physical dependence on

and tolerance to morphine in the rat. Dmg Res., 27, 1676- 1679.

KOYUN~UOGLU,H., KEYER-UYSAL,M., BERKMAN, K., GONGOR, M. & Gmç,E.

( 1979). The relationship between morphine, aspartic acid and L-asparaginase in

rats. Eur. J. Phunnacoi., 60, 369-3 72.

KREBS,H.A. (1935). Metabolism of amino acids. IV. Synthesis of glutamine fiom

glutamic acid and ammonia, and the enzymic hydrolysis of glutamine in animal

tissue. Biochem. J., 29, 195 1- 1969.

KUHAR, M.J., PERT,C.B. AND SNYDER,S.H. (1973). Regional distribution of opiate

receptor binding in monkey and human brain. Nature (Land), 245, 447-450.

K~SUWADA,T., KASHIWABUCHI, N., MORI, H., SAKIMURA,K., KUSHIYA,E.,

ARAKI, K., MEGURO,H., MASAKI,H., KIMANISHI, T., ARAKAWA,M. & MISHINA,

M. (1 992). Molecular diversity of the NMDA receptor channel. Nature, 358, 36-

41. LAMBERT,D.G. ( 1993). Signal transduction: G proteins and second messengers. Br.

J. Anaes., 71, 86-95.

LAMBERT, D.G., ATCHESON,R., HIRST,RA. & ROWBOTHAM,D.J. (1993). Effects of

morphine and its metabolites on opiate receptor binding, CAMPformation and

[3 Hlnoradrenaline release fiom SH-SYSY cells. Biochem. Phannacol., 46, 1 145-

1 150.

LAMBERT,D.G., WOJCIKIEWICZ,R.J.H., SAFRANY,S.T., WHITHAM,E.M. &

NAHORSKI,SR. (1992). Muscrinic receptors, phosphoinositide metabolism and

intracellular calcium in neuronal cells. Prog. Nezcro-Psychophamcol. & Biol.

Psychiatry, 16, 253-270.

LARSON,A.A., VAUGHT,J.L. AND TAKEMORI,A.E. (1980). The potentiation of

spinal analgesia by leucine enkephalin. Eur. J. Phannacol., 6 1, 3 8 1-3 83.

LASCHKA,E., TESCEMACHER,W., MEHRAEIN, P. & HERZ, A. (1976). Sites of

action of morphine involved in the development of physical dependence in rats.

Psychopharmacologia (Berl..), 46, 14 1- 147.

LAW, P.Y. & LOH, H.H. (1986). Cell-ffee desensitization of opioid inhibition of

adenylate cyclase in neuroblastoma x glioma (NG108- 15) hybrid ce11 membranes.

J. Neurochem., 47,733-737.

LEACH, R.P.,SHEARS, S.B., KRK, C.J. & TITHERADGE,M.A. (1986). Changes in fiee

cytosolic calcium and accumulation of inositol phosphates in isolated heptatocytes

by [leulenkephalin. Biochem. J., 238, 537-542. LEE, N.M., LEYBIN,L., CHANG,J.-K. AND LOH, H.H.(1980). Opiate peptide

interaction: Effect of enkephalins on morphine analgesia. Eur. J. Phamcol., 68,

181-185.

LEE,P.H., MCNUTI, R. W. AND CHANG,K.J. (1 993). A nonpeptidic delta opioid

receptor agonist, BW373U86, attenuates the developrnent and expression of

morphine abstinence precipitated by naloxone in rat. J. Phannucol. Exp. Ther.,

267, 883-887.

LEMBERGER,L., & RUBIN.A. ( 1976). chapter 5 : Morphine and Morphine

Substitutes. In Physiological Dipsilion of Drugs of Abuse. New York:

S pectmm Publications Inc.

LI, C.H. AND CHUNG,D. (1976). Isolation and structure of an untriakontapeptide

with opiate activity from came1 pituitary glands. Proc. Nd.Acad Sei., 73, 1145-

1148.

LI, P.P.,SIBONY, D., GREEN,M.A. & WARSH,J.J. (1993). Lithium modulation of

phosphoinositide signaiing system in rat cortex: Selective effect on phorbol ester

binding. J. Neurochern., 61, 1722- 1730.

LING, N., BURGUS,R. & GUILLEMIN,R. (1976). Isolation, primary structure, and

synthesis of alpha-endorphin and gamma-endorphin, two peptides of hypothalamic-

hypo p hy sial origin with rnorphinornimetic activity . Proc. Nutl. Acad LISA, 73,

3042-3046. LOWE, D. (1992). Concept of dBerent receptors and transduction mechanisms for

excitatory amino acids. In Excitatory Amino Acidsr Fidia Research Foundation

Symposizrm Series, VOL9, Simon, R.P.,Ed., Thieme Medical Publishers, Inc., New

York, 11 1-1 12.

MACDERMOTT,A.B., MAYER,M.L., WESTBROOK,G.L., SMITH, S.J. & BARKER,

J. L ( 1986). NMDA-receptor activation increases cytoplasrnic calcium

concentration in cultured spinal cord neurones. Nature, 321, 5 19-522.

MALDONADO,R., NEGUS,S. AND KOOB, G.F. (1992). Precipitation of morphine

withdrawal syndrome in rats by administration of mu-, delta- and kappa-selective

opioid antagonists. NeurophannacoL, 31, 1231-1241.

MALDONADO,R., VALVERDE,O., GARBAY, C. & ROQUES,B.P. (1995). Protein

kinases in the locus coemleus and periaqueductai gray matter are involved in the

expression of opiate withdrawai. Nmnyn-Schmiedeberg S Arch. PharrnacoL , 352,

565-575.

MANJ~,H.K., ETCHEBERRIGARY,R., CHEN, G. & OLDS, J.L. (1993). Lithium

decreases membrane-associated protein kinase C in hippocarnpus: Selectivity for

the a isozyrne. J. Neurochem., 61,2303-23 10.

WSOUR,A., FOX, C.A., AKL, H.& WATSON,S.J. (1995). Opioid-receptor rnRNA

expression in the rat CNS: anafornical and hnctional implications. Trent&

Nezrrosci., 18, 22-29. MANZONI, O.J. J., FINIELS-MAEUIER,F., SASSETII,I., BLOCKAERT,J., LE PEUCH,C

& SLADECZEK, F.A.J. (1990). The glutamate receptor of the Qp-type activates

protein kinase C and is regulated by protein kinase C. Neurosci Lett., 109, 146-

151.

MAO, J., PRICE,D.D. & MAYER,D.J. (1994). Thermal hyperalgesia in association

with the development of morphine tolerance in rats: Role of excitatory amino acid

receptors and protein kinase C. J. Neurosci., 14, 230 1-23 12.

MAO, J., PRICE,D.D. & MAYER, D.J. (1 995). Mechanisms of hyperalgesia and

morphine tolerance: a current view of their possible interactions. Pain, 62, 259-

274.

MAO, J.R., PRIcE, D.D., PHILLIPS,L.L., LU, J. & MAYER,D.J. (1995). Increases in

protein kinase C gamma irnrnunoreactivity in the spinal cord of rats associated with

tolerance to the anaigesic efTects of morphine. Brain Res., 677, 257-267.

M~K,P.,BEN-ELIYAHU, S., GOLD, M. & LIEBESKIND,J.C. (1991 a). Excitatory

arnino acid antagonists (kynurenic acid and MK-80 1) attenuate the development of

morphine tolerance in rats. Bain Res.. 547, 77-8 1.

WK,P., BEN-ELIYAHU, S., VACCARINO,A.L. & LEBESKIND, J-C. (199 lb).

Delayed application of MK-80 1 attenuates development of morphine tolerance in

rats. Brain Res., 558, 163-165. MARKS, F., HANKE, B., THASTRUP,O. & FURSENBERGER,G. (1991). Stimulatory

effect of thapsigargin, a non-TPA-type turnor prornoter, on arachidonic acid

metabolism in the murine keratinocyte line HEL30 and on epidermd ce11

proliferation Ni vivo as compared to the effects of phorbol ester TPA.

Carcinogenesis, 12, 149 1-1497.

MARTIN,W. R., EADES,CG., THOMPSON, J. A., HUPLER,R.E. AND GILBERT,P. E.

(1 976). The effects of morphine- and nalorphine-like drugs in the nondependent

and morphine-dependent chronic spinal dog. J. Pharmacol. m. Ther., 197, 5 17-

53 2.

MASU,M., NALUMA, Y., MORIYOSHI,K., ISHII,T., AKAZAWA, C. & NAKANISHI,S.

( 1994). Molecular characterization of NMDA and metabotropic glutamate

receptors. Ann. NYAcad Sci., 707, 153- 164.

MASU, M., TANABE,Y., TSUCHIDA,K., SHIGEMOTO,R., & NAKANASHI,S. (199 1).

Sequence and expression of a metabotropic glutamate receptor. Nahre, 349, 760-

765.

~~TSUOW,I., MALDONADO, R., DEFER, N., NOEL, F., HANOUNE, J. & ROQUES.

B.P.(1994). Chronic morphine administration causes region-specific increase of

brain type VIII adenylyl cyclase mRNA. Eur. J. Pharmacoi., 268, 2 1 5-22 1.

MAYER,D.J., MAO, J. & PRICE,D.D. (1995). The development of morphine

tolerance and dependence is associated with translocation protein kinase C. Pub,,

61, 365-374. MAYER,M.L. & WESIBROOK, G.L. (1987a). The physiology of excitatory amino

acids in the vertebrate central nervous syaem. Prog. Neurobiol., 28, 197-276.

MAYER, M.L. & WESTBROOK,G.L. (198%). Permeation and block of N-rnethyl-D-

aspartic acid receptor channels by divalent cations in mouse cultured centrai

neurones. J. Physiol., 394, 50 1-527.

MAYER,M.L., MACDERMOTT,AB., WESTBROOK,G.L., Sm,S. J. & BARKER,

J.L ( 1987). Agonist- and voltage-gated calcium entry in cultured mouse spinal

cord neurons under voltage clamp measured using Arsenazo UI. J. Neurosci., 7,

3 23 0-3244.

MAYER, M.L., WESTBROOK,G.L. & GUTHRIE, P.B.(1 984). Voltage-dependent

block by M~~+of NMDA responses in spinal cord neurones. Nature, 309, 26 1-

263.

PUIEGURO, H., MORI, H., ARAN, K.,KUSHIYA, E.,KUTSUWADA, T., YAMAZAIU, M.,

KUMAMSHI, T., ARAKAWA, M., S- K. & MJSHINA,M. (1992).

Functional characterization of a heteromeric NMDA receptor channel expressed

from cloned cDNAs. Mure, 357, 70-74.

MESTEK,A., HURLEY,J.H., BYE, L.S.,CAMPBELL, A.D., Cm, Y., TIAN, M., LIU, J.,

SCHULMAN,H. & YU. L. (1995). The human p opioid receptor: Modulation of

functionai desensitization by calcium~calmodulin-dependentprotein kinase and

protein kinase C. J. Neurosci., 15, 2396-2406.

MILLER,R. J. (1 99 1). The revenge of the kainate receptor. Trends Neurosci, 14,

477-479. MIYAMOTO,Y., PORTOGHESE,P.S. & TAKEMORI,A.E. (1993). Involvement of deltaz

opioid receptors in the development of morphine dependence in mice. J.

Phannacol. Exp. Ther., 264, 1 14 1- 1 145.

MONAGHAN,D.T., BRIDGES,R.J. & COW, C.W. (1989). The excitatory arnino

acid receptors: Their classes, pharmacology and distinct properties in the funetion

of the central nervous systern. Am. Rev. Phannacol. ToxicoL, 29, 365 - 402.

MONYER,H., SPRENGEL,R., SCHOEPFER,R., HERB,A., HIGUCHI,M., LOMELI,H.,

BURNASHEV,N., SAKIULANN,B. & SEEBURG, P.H. (1992). Heteromenc NMDA

receptors: Molecular and functiond distinction of subtypes. Science, 256, 12 17-

1221.

MORIYOSHI,K., MASU, M., ISHIIKT.,SHIGEMOTO, R., MIZUNO,N., & NAKANISHI,S.

( 199 1 ). Molecular cloning and characterization of the rat NMDA receptor.

Nature, 354, 3 1-3 7.

MOSKOWTZ,A.S. AND GOODMAN,R.R. (1985). Autoradiographic distribution of

mu1 and mu2 opioid binding in the mouse central nervous system. Brain Res..

360, 1 17- 129.

MUCHA, R.F. AND b-lLANT, H. (1 980). Log dosdresponse curve flattening in rats

after daily injection of opiates. Psychopharmacof., 7 1, 5 1-61 .

MUCHA, R.F., KALANT, H. AND LINSEMAN,M.A. (1979). Quantitative relationships

among measures of morphine tolerance and physical dependence in the rat.

Pharmacol., Biochem. & Behav., lO,397-405. MLIRPHY,S.N. & MILLER,R. J. (1989). Regulation of ~a*influx into stnatal

neurons by kainic acid. J. Phmacol. m. mer., 249, 184- 193.

NAUJTMA, Y., IWAKABE, H.,AKAZAWA, C., NAWA,H., SHIGEMOTO, R., MUUNO,

N. & NAKANISHI,S. ( 1993). Molecular characterization of a novei retinal

metabotropic glutamate receptor mGluR6 with a high agonist selectivity for L-2-

arnino-4-phosphonobutyrate. J. Biol. Chem., 268, 868-873.

NAKANISHI,S. (1 992). Molecular diversity of glutamate receptors and implications

for brain fbnction. Science, 258, 597-603.

NNCANISHI, S. (1 994). The molecular diversity of glutamate receptors. Mofendar

Nezrrobiology: Mechanisms Cornmon IO Brain, Skin, and Immune Syslem, 8 5 -98.

NAKANISHI,S., NOLIE,A., KITA, T., N-, M., CHANG,A.C.Y., COHEN,S.N.

& NUMA,S. (1979). Nucleotide sequence of cloned cDNA for bovine

corticotropin-beta-lipotropin precursor. Nalirre, 278, 423-427.

NARITA,M., FENG,Y., M., HOSKINS,B. AND HO, I.K. (1994a). A

protein kinase inhibitor, H-7, inhibits the development of tolerance to opioid

antinociception. Etw. J. Phmacof., 27 1, 543-545.

NARITA,M., MAKIMURA, M., FENG, Y.,HOSKINS, B. & HO, I.K.(1 994b). Influence

of chronic morphine treatment on protein kinase C activity: cornparison with

butorphanol and implication for opioid tolerance. Brain Res.. 650, 175- 179.

neurones. J. PhysioL, 394, 501-527. NICHOLSON,D., RED, A. & SAWYNOY1. (1991). Effects of forskolin and

phosphodiesterase inhibitors on spinal antinociception by morphine. Pharmacol.

Biochem. & Beh,38, 753-758.

NICOLEITI,F.. BRUNO,V., COPANI,A., CASABONA,G. & WOPFEL, T. (1996).

Metabotropic glutamate receptors - a new target for the therapy of

neurodegenerative disorders. Trends Neurosci., 19, 267-27 1.

NICOLETT, F., CONDORELLI,D.F., DELL'ALBANI, P., AROMCA,E., CATANIA, M.V.,

SORTINO.M.A. & REYMANN, KG.(1991). Metabotropic glutamate receptors

and neu ro pl ast ici ty . In Excitatory Amino Acids, Fidia Reseurch Foundarion

Symposium Series, Vol. 9, R.P. Simon, Ed. (Thieme Medical Publishers, New

York), pp. 167-1 70.

NIsHIMURA. S.L., RECHT, L.D. AND PASTERNAK,G. W. (1 984). Biochemical

3 characterization of high affnity H-opioid binding: further evidence for mu, sites.

Mol. Pharmacol. 25, 29- 37.

NODA,M., FURUTANI,Y., TAIWWSHI,H., TOYOSATO,M., HIROSE, T., ~NAYAMA,

S., NAKANISHI,S. & NUMA, S. (1 982). Cloning and sequence analysis of cDNA

for bovine adrenal preproenkephalin. Nu~ure,295, 202-206.

NOWAK,L., BREGESTOVSKI,P., ASCHER, P., HERBERT,A. & PROCHIANTZ,A.

( 1984). Magnesium gates glutamate-activated channels in mouse central neurones.

Nalzcre, 307, 462-465. OH, K.W.,MAKIMURA, M., JAW, S.P., HOSKINS,B. & HO, I.K. (1992). Effects of

beta-funaitrexarnine on butorphanol dependence. Phmacol. Biochem. & Behav.,

42, 29-34.

OKAJMA, F., ToMURA, H.& KONW, Y. (1993). Enkephaiin activates the

2- phospholipase ClCa system through cross-talk between opiate receptors an P,--

purinergic or bradykinin receptors in NGIO8- 15 cells: A permissive role for

pertussis toxin sensitive G-proteins. Biochem. J., 290, 24 1-247.

OKAMOTO,N., HORI, S., AKAZAWA, C., HAYASM,Y., SHIGEMOTO, R., MIZUNO,N.

& NAKANISHI,S. (1 994). Molecular characterization of a new rnetabotropic

glutamate receptor mGluR7 coupled to inhibitory cyclic AMP signal transduction

J. Biol. Chern., 269, 1231-1236.

PALMER, E., MONAGHAN,D.T. & COTMAN, C.W.(1989). Tram-ACPD, a selective

agonist of the phosphoinositide-coupled excitatory amino acid receptor. Eur. J.

Pharmacol., 166, 585-587.

PASTERNAK. G. W. AND WOOD,P.J. (1986). Multiple mu opiate receptors. Lrfr Sci.,

38, 1889- 1898.

PAUL, D.,BODNAR, R.J., GISTRAK, M.A. AND PASTERNAK,G. W. (1 989). Different p

receptor subtypes mediate spinal and supraspinal anaigesia in mice. Eur. J.

Pharmacol., 168, 307-3 14.

PAXINOS, G. & WATSON,C. (1986). The rat brain in stereotôuc coordinates.

Academic Press Inc., San Diego, California. PELLEGRNI-GWETRO,D.E., RUGGIERO,M., GIANNELLI,S., CHIARUGI, S. &

MOROM,F. (1988). Morphine withdrawal in vitro: potentiation of agonist-

dependent polyphosphoinositide breakdown. Eur. J. Phmc~coi.,149, 297-306.

PERCHELLET, E.M., GALI,H.U., GAO,X.M. & PERCHELLET,J.P. (1 993). Ability of

the non-phorbol ester-type tumor-promoter thapsigargin to mimic the stimulatory

effects of 12-O-tetradecanoylphorbol-13-acetate on ornithine decarboxylase

activity, hydroperoxide production, and macromolecule synthesis in mouse

epidermis in vivo. IntematII J. of Cancer, 55, 1O3 6- 1043.

PEMYASAMY,S. & HOSS,W. (1990). Kappa-opiate receptors stimulate

phosphoinositide turnover in rat brain. Life Sci., 47, 2 L 9-225.

Pm, J.-P. & DWIOSN, R. (1995). Review: Neurotransmitter receptors 1. The

Metabotropic glutamate receptors: Structure and Functions. NeuropharmacoL,

34, 1-26.

PORRECA, F., HEYMAN,J.S., MOSBERG,I., OMNAAS,J.R. AND VAUGHT,J.L. (1987).

Role of mu and delta receptors in the supraspinai and spinal anaigesic effects of

p-~enz,~-~en~]enke~halin in the mouse. J. Pharmacol. Exp. Thet-.,241, 393-

400.

PORTOGHESE, P.S., NAGASE,H., MALONEYHUSS, K.E., LN, C.-E. &

TAKEMORI,A.E. (199 1). Role of spacer and address components in

peptidomimetic 6 opioid receptor antagonists related to naitrindole. J. Meu!

C'hem., 34, 1715-1720. PRZEWLOCKA,B., DZIEDZICKA, M., LASON, W. & PRZEWLOCKI,R. (1 992).

Differential effects of opioid receptor agonists on nociception and CAMP level in

the spinal cord of monoarthritic rats. Life Scî., 50, 45-54.

RAFFA, R.B. & MARTINU, R.P. (1992). Morphine antinociception is mediated

through a LiCI-sensitive, IP3- restorable pathway. Eur. J Phmacoi., 2 15, 3 57-

358.

RAFFA, R.B., CONNELLY,C.D. & MAR-, R.P. (1992). Opioid efficacy is linked to

the Lithium chloride-sensitive inositol- 1,4,5-tnsphosphate-restorable pathway.

Eztr. J. Pharmacoi., 2 17, 22 1 -223.

RAMABADRAN, K. (1 985). An analysis of precipitated withdrawal in rats acutely

dependent on morphine. Jap. J. Phmacol., 37, 307-3 16.

RASMUSSEN,H. (1986a). The calcium messenger system (First of two parts). New

Engl. J. Med, 314, 1094-1 101.

RASMUSSEN,H. (1986b). The calcium messenger system (Second of two parts). New

Engl. J. Med, 314, 1164-1170.

RASMUSSEN,K., FULLER, R.W., STOCKTON, M.E., PERRY,K.W., SWINFORD, R.M. &

ORNSTEM,P.L. (199 la). NMDA receptor antagonists suppress behaviors but not

norepinephrine turnover or locus coemleus unit activity induced by opiate

withdrawal. Eur. J. Phannacol., 197, 9-16.

RASMUSSEN, K.,KRYSTAL, J.H. & AGHAJANIAN,G.K. (1 99 I b). Excitatory amino

acids and morphine withdrawal: difEerentiai effects of centrai and peripheral

kynurenic acid administration. Psychophannacoiogy, 105, 508-5 12. REISNE, T. & BELL,G.I. (1993). Molecular biology of opioid receptors. Trends

Neurosci., 16, 506-510.

RIEDEL,G. & REYMANN, K. (1993). An antagonist of the metabotropic glutamate

receptor prevents LTP in the dentate gynis of freely moving rats.

Neirrophamacob . 32, 929-93 1.

ROTHMAN,R.B. & WESTFW, T.C. (1982). Aiiostenc coupling between morphine

and enkephalin receptors in vitro. Mol. Phmacof., 21, 548-557.

SALT,T.E. & EATON,S.A. (1995). Distinct presynaptic metabotropic receptors for L-

AP4 and CCGI on GABAergic terminais: Pharmacological evidence suing novel

a-methyl derivative mGluR antagonists, MAP4 and MCCG, in the rat thalamus it~

vivo. Netirosci., 65, 5-13.

SAUGSTAD,J.A., KINzIE, J.M., MULVMILL,E.R., SEGERSON,T.P. & WES~ROOK,

G.L. (1 994). Cloning and expression of a new member of the L-2-arnino-4-

phosphonobutyric acid-sensitive class of metabotropic glutamate receptors. Moi.

Pharmacol., 45, 367-3 72.

SCHILLER,P. W., NGUYEN,T.M.-D., WELTROWSKA,G., WILKES,B.C., MARSDEN,

B.J., LEMIEUX,C. & CHUNG,N.N. (1992). Differential stereochemical

requirements of p vs. 6 opioid receptors for ligand binding and signai transduction:

development of a class of potent and highly ô-selective peptide antagonists. PTOC.

Natl. Acad Sci. USA, 89, 1 1871-1 1875. SCHILLER,P.W., WELTROWSKA, G., NGUYEN,TM.-D., WILKES,B.C., CHUNG,N.N.

& LEMIEUX,C. (1993). TTPP[y]: a highiy potent and stable pseudopeptide 6

opioid receptor antagonist with extraordinary 6 selectivity. J. Med, Ch.,36,

3 182-3 187.

SCHOEPP,D.D. & CONN, P.J. (1993). Metabotropic glutamate receptors in brain

fùnction and pathology. Trenak Phmacol. Sci., 14, 13-25.

SCHOEPP,D.D., GOLDSWORTHY, J., JOHNSON,B.G., SALHOFF,C.R. & BAKER,S.R.

( 1994). 3,s-DihydroxyphenyIgiycineis a highly selective agonist for

phosphoinositide-Iinked metabotropic glutamate receptors in the rat hippocampus.

J. Nezmchem., 63, 769-772.

SCHOEPP,D.D., JOHNSON,B.G. & MONN,J.A. (1992). Inhibition of cyclic AMP

formation by a selective metabotropic glutamate receptor agonist. J. Nezîrochem.,

58, 1 184- 1 186.

SCHOEPP,D.D., JOHNSON,B.G., SALHOFF, C.R., MCDONALD, J.W. & JOHNSON,

M.V. (199la). In vitro and in vivo pharmacology of pans- and cis-(5)-I-amino-

1,3-cyclopentanedicarboxylic acid: dissociation of metabotropic and ionotropic

excitatory amino acid receptor effects. J. Neicrochem., 56, 1789- 1796.

SCHOEPP,D.D., JOHNSON,B.G., SMITH, E.C.R.& MCQUAID,L.A. (1 990). Stereo

selectivity and mode of inhibition of phosphoinositide-coupled excitatory amino

acid receptors by 2-amino-3-phosphonopropanoic acid. Mol. Phannacol., 38, SCHOEPP,D.D., JOHNSON,B.G., TRUE,R.A. & MONN,J.A. (1% lb). Cornparison of

( 1S,3R)- 1-arninocyclopentane- 1,3-dicarboxylic acid (lS,3R -ACPD)- and 1R,3S-

ACPD-stimulated brain phosphoinositide hydrolysis. Eur. J. Pharmacol., 207,

35 1-353.

SCHOFFELMEER,A.N., RICE, K.C., JACOBSON,A.E., VANGELDERON, J.G.,

HOGENBOOM,F., HENA, M.H. & MULDER,A.H. (1988). Mu-, delta- and kappa-

opioid receptor-rnediated inhibition of neurotransmitter release and adenylate

cyclase activity in rat brain slices: studies with fentanyl isothiocyanate. Ew. J.

Pharmacol., 154, 169-178.

SCHOFFELMEER,A.N., VERHEIJDEN,P.F., HOGENBOOM,F., MULDER,A.H & STOOF,

I.C. (1987). Opioid and D-2 receptor mediated inhibition of forskolin-stimulated

cyclic AMP efflux from rat neostriatal slices. Neuropharmacol., 26, 785-787.

Sm,S.K., NIRENBERG,M. & KLEE, W.A. (1 975). Morphine receptors as

regulators of adenylate cyclase activity. Proc. Nd. Acad Sci USA, 72, 590-594.

SIIVIANTOV, R. AND SNYDER,S.H. (1976). Morphine-like peptides in rnamrnalian

brain: Isolation, structure elucidation, and interactions with the opiate receptor.

Proc. Nad. Acad Sci. USA, 73,2525-2519.

SLADECZEK,F., PIN, J.P., RECASENS,M., BOCKAERT,J. & WEISS,S. (1 985).

Glutamate stimulates inositol phosphate formation in striatal neurones. Nature.

317, 717-719. SMART,D. & LAMBERT, D.G. ( 1996). The stimulatory effects of opioids and their

possible role in the development of tolerance. Trenak Phamacof. Sci., 17, 264-

269.

SWT, D., SmG. & LAMBERT,D.G. (1 994). mu-Opioid receptor stimulation of

inositol ( 1,4,5) trisphosphate formation via a pertussis toxin-sensitive G protein. J.

Neurochem., 62, 1009- 10 14.

SMART,D., SMITH, G. & LAMBERT,D.G. (1995). p-Opioids activate phospholipace

C in SH-SYSY neuroblastoma cells via calcium-channe1 opening. Biochem. J.,

305, 577-582.

SOFUOGLU,M., PORTOGHESE,P. S. & TAKEMORI,A.E. (1 993). 7-

Benzylidenenaltrexone (BNTX): a selective delta 1 opioid receptor antagonist in

the mouse spinal cord. Lije Sci., 52, 769-775.

SOMMER,B. & SEEBURG.P.H. (1992). Glutamate receptor channels: novel

properties and new clones. Trends Phannacol. Sci., 13, 29 1-296.

S-IEPPUHN,KG. & TURSKI,L. (1993). Diazeparn dependence prevented by

glutamate antagonists. Proc. Naff.Acad Sci. (USA), 90, 6889-6893.

STEVENS,C.W. AND YAKSR T.L. (1989a). Magnitude of opioid dependence after

continuous intrathecal infusion of p- and 3-selective opioids in the rat. Eur. J.

Pharmacol., 166, 467-472.

SEVENS, C.W. AND YAKSH,T.L. (1989b). Time course characteristics of tolerance

development to continuously infused antinociceptive agents in rat spinal cord. J.

Phamcol. Exp. mer., 251(1), 2 16-223. STEWART,P.E. AND HAMMOND,DL. (1 993). Evidence for delta opioid receptor

subtypes in rat spinai cord: Studies with intrathecal naltriben, cyclicp-l'en2, D-

~en5]enke~halinand p41a2, ~lu~]deltor~hin.J. Phmacol. Exp. 7?zm,266,

820-828.

SUGIYAMA,H., ITO. 1. & H~EzONO,C. (1987). A new type of glutamate receptor

linked to inositol phospholipid metabolism. Nature, 325, 53 1-533.

SUH,H.H. AND TSENG,L.F. (1990a). Different types of opioid receptors mediating

analgesia induced by morphine, DAMGO, DPDPE, DADLE and beta-endorphin in

mice. Nmnyn-Schmied Arch. Phannacol., 342, 67-7 1.

Sm, H.H.AND TSENG,L.F. (1990b). Lack of antinociceptive cross-tolerance

between intracerebroventriculariy administered P-endorphin and morphine or

DPDPE in mice. Life Sci., 46, 759-765.

TAKEMORI,A.E. & PORTOGHESE,P.S. (1 987). Evidence for the interaction of

morphine with kappa and delta opioid receptors to induce analgesia in B-

fùnaltrexamine-treated rnice. J. Phannacoi. Erp. î"her.,243, 9 1-94.

TAKEMORI,A.E., LOH, H.H. AND LEE,N.M. (1993). Suppression by dynorphin A and

1 [Des-Tyr Idynorphin A peptides of the expression of opiate withdrawai and

tolerance in morphine-dependent rnice. J. Phurmacol. Exp. mer., 266, 12 1- 124.

TANGANELLI,S., ANTONELLI,T., MORARI, M., BIANCHI,C. & BEANI,L. (1991).

Glutamate antagonists prevent morphine withdrawal in mice and guinea pigs.

Neurosci. Lett., 122, 270-272. TATUM,A.L., SEEVERS,M.H. & COLLLNS,K.H. (1929). Morphine addiction and its

physiological interpretation based on experimental evidences. J. Pharmacoi. Erp.

Ther., 36, 447-475.

TERENTUS,L. (1973). Characteristics of the "receptor" for narcotic analgesics in

synaptic plasma membrane fraction fiom rat brain. Acta Phannacofogica et

Toxicologica, 33, 3 77-3 84.

TERWILLIGER,R.Z., BE~R-JOHNSON, D., SEVARINO, KA., Cm,S.M. &

NESTLER,E.J. (199 1). A general role for adaptations in G-proteins and the cyciic

AMP system in mediating the chronic actions of morphine and cocaine on neuronal

fiinction. Brain Res., 548, 100-1 10.

THASTRUP,O., CULLEN,P.J., DROBAK, B.K.,HANLEY, M.R. & DAWSON,A.P.

(1 990). Thapsigargin, a tumor promoter, discharges intracellular ca2+stores by

specific inhibition of the endoplasrnic reticulum Ca2(+)-ATPase. Proc. Nd.

Acud Sci. (USA), 87, 2466-2470.

THOMSEN,C., BOEL,E. & SUZDAK,P.D. (1994). Actions of phenylglicine analogs at

subtypes of the metabotropic glutamate receptor family. Eur. J. Pharmacol., 267,

77-84.

TISEO.P.J. & LNTUEWSI, C.E. (1 993). Attenuation and reversal of morphine tolerance

by the cornpetitive N-methyl-D-aspartate receptor antagonist, LY2746 14. J.

Pharmacol. Exp. 7?ier.,264, 1090- 1096. TISEO,P.J. & YAKSH,TL. (1993). Dose-dependent antagonism of spinal opioid

receptor agonists by naloxone and naltrindole: additional evidence for 3-opioid

receptor subtypes in the rat. Eur. J. Phannacoi., 236, 89-96.

TRUJTLLO,K. A. & AKIL, H. (1991). Inhibition of morphine tolerance and dependence

by the NMDA receptor antagonist MK-80 1. Science. 251, 85-87.

TRUJILLO, K.A. & AKL, H. (1994). MUbition of opiate tolerance by non-cornpetitive

N-methyl-D-aspartate receptor antagonists. Brain Res., 633, 178- 188.

Tsu, R.C., CHAN, J.S.C. & WONG,Y.H. (1995). Regulation of multiple effectors by

the cloned delta-opioid receptor - stimulation of phospholipase C and type II

adenylyl cyclase. J. Neurochem., 64,2700-2707.

UEDA,H., M~YAMAE,T., HAYASHI,C,, WATANABE,S., FUKUSHIMA,N., SASAKI,Y.,

IWAMURA, T. & MISU,Y. (1995). Protein kinase C involvement in homologous

desensitization of 6-opioid receptor coupled to Gii-phospho1ipaseC activation in

Xet~opu'ioocytes. J. Nezwosci., 15, 7485-7499.

VALENTINO,R. J. & WEHBY,R.G. (1989). Locus coemleus discharge characteristics

of morphine-dependent rats: Effects of naltrexone. Brain Res., 488, 1 26- 1 34.

VAN VLIET,B.J., VANRIJSWIJK, A.L., WARDEH,G., MULDER,A.H. &

SCHOFFELMEER,AN. (1 993). Adaptive changes in the number of Gs- and Gi-

proteins underlie adenylyl cyclase sensitization in morphine-treated rat stnatal

neurons. Eur. J. Phamtacol., 245, 23-29. VAUGHT,J.L. AND TAKEMORI,A.E. (1979). A further characterization of the

differential effects of leucine enkephalin, methione enkephalin and their analogs on

morp hine-induced analgesia. J. Phannucol. Exp. mer., 2 11, 280-283.

VAUG~,J.L., ROTHMAN,R.B. AND WESTFALL,T.C. (1982). Mu and delta receptors:

Their role in analgesia and in the differential effects of opioid peptides on

analgesia. Life Sci., 30, 1443- 145 5.

WANG, J.F., REN,M.F., XUE, J.C. & HAN, J.S. (1 993). Cyclic AM.mediates mu and

delta, but not kappa opioid anaigesia in the spinal cord of the rat. Life Sci., 52,

1955-1960.

W~ms,J. C. & COLLINGRIDGE,G.L. (1994). Phenylglycine denvatives as

antagonists of metabotropic glutamate receptors. Trends Pharmacol. Sei., 15, 3 3 3 -

342.

WIDNELL,K.L., RUSSELL, D.S. & NEsTLER, E.J. (1994). Regulation of expression of

CAMP response element-binding protein in the locus coeruleus in vivo and in a

locus coeruleus-like ce11 line in vitro. Proc. NatL Acad Sci USA, 91, 10947-

1095 1.

WOLLEMANN,M., BENYHE,S. AND SIMON,J. (1993). The kappa-opioid receptor:

Evidence for the different subtypes. Life Sci., 52, 599-6 1 1.

WOLOZiN, B.L.AND PASTERNAK,G.W. (198 1). Classification of multiple morphine

and enkephalin binding sites. Proc. Natl. Acad Sci. (USA), 78, 6 1 8 1- 6 1 8 5. WONG,E.H.F., 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. Proc. Nuti. Ac&. Sci. (USA), 83, 7104-7108.

YAMAMOTO,T., OHNO, M. AND UW, S. (1988). A selective kappa-opioid agonist,

U-50,488H, blocks the development of tolerance to morphine analgesia in rats.

Ezrr. J. Pharmacol. , 156, 17 3 - 176.

YASUDA,K., RAYNOR, K.,KONG, H., BREDER,C., TAKEDA,J., ~ISINE,T. & BELL,

G.I. (1993). Cloning and functional cornparison of K and 6 opioid receptors fiom

mouse brain. Proc. Natl. Acad Sci. USA, 90, 6736-6740.

YOBURN,B.C., LUTFY, K., SIERR~ V. AND TORTELLA,F.C. (1990). Tolerance

develops to spinal morphine anaigesia but not morphine-induced convulsions. Eirr.

J. Pharmacol., 176,63-67.

YOUNG,E.A., NEFF, N.H. & JIC CONSTANT IN OU, M.(1994). Phorbol ester

administration transiently increases aromatic L-amino acid decarboxylase activity

of the mouse striatum and midbrain. J. Nezdrochern., 63, 694-697.

YOUNG,GA. AND KHAZAN, N. (1985). Cornparison of abstinence syndromes

following chronic administration of mu and kappa opioid agonists in the rat.

Pharmaod., Biochem. & Behav., 23,457460.

YU, V.C. & SADEE,W. (1 986). Phosphadidylinositol turnover in neuroblastorna cells:

regulation by bradykinin, acetylcholine, but not mu- and delta-opioid receptors.

Neurosci. Lett., 7 1, 2 19-223. YU,V.C., EIGER,S., DUAN,D.S., LAMEH, J. & SADEE,W. (1990). Regdation of

cyclic AMP by the mu-opioid receptor in human neuroblastoma SH-SYSY cells.

J. Neurochem., 55, 1390-1396.

ZHANG,A.-Z. AND PASTERNAK,G.W. (1980). Mu and delta opiate receptors:

correlation with high and low afFuity opiate binding sites. Eur. J. Pharmacol.. 67,

323- 324.

ZHANG.L. J., WANG, X.J. & HAN,J.S. (1990). Phorbol ester suppression of opioid

analgesia in rats. Llfe Sci., 47, 1775- 1 782. APPENDIX 1: ORIGINAL CONTRIBUTIONS Original Contributions

We are the first to examine the role of metabotropic glutamate receptors

(mGluRs) in the development of morphine tolerance and dependence. Although other investigators hypothesized a role for protein kinase C (PKC) in opioid tolerance and dependence, they used non-selective protein kinase inhibitors, and thus it was not entirely clear whether the results they obtained were due to PKC,or to other protein kinases which these agents affected. Thus, we were the fkst to show that highly selective inhibition of PKC activation attenuates the development of morphine dependence. Moreover, we were the iirst to demonstrate that selective inhibition of intracellular ca2' release also attenuated morphine dependence. Although other investigators examined the role of 6-opioid receptors in the development of morphine tolerance and de pendence, t hey used antagonists which also have significant antagonistic activity at p- and K-opioid receptors, rendenng their results inconclusive.

We thus estabiished a specific role for 8-opioid receptors by using antagonists (TIPP and TLPP[w]) which are highly selective for 6-opioid receptors, and show no activity at p- or K-opioid receptors. Furthemore, we proposed a mode1 which, although it incorporates other models, extends the possibilities to a possible interaction of opioid receptors and rnGluRs, via actions on intracellular second messengers. Contributions on CO-authoredpapers

Dr. Terence J. Coderre is a CO-authoron al1 of the manuscripts because he is my thesis supervisor and played a vitai role in the completion of my doctoral research and dissertation. The manuscript contained in Chapter 3 is also CO-authoredby

Jennifer Ritchie because she provided vaiuable aid in data collection when I was pregnant. The manuscript contained in Chapter 6 has several CO-authors.Michelle

Shapiro was an undergraduate Honours student who aided in the collection of data under my direct supervision. Peter Schiller and Grazyna Weltrowska are members of

Dr. Peter Schiller's laboratory, where TIPP and TIPP[v] were synthesized. APPENDIX II: COPYRIGHT PERMISSION

MF;U11ILLRN PESS LTD 10-JLPI-1596 16:5t . .... S..' . .._ 1 . . , . .-.. :...... * >

Insüfut de..%cherches . cliniques de Montréai ~[i~i&'~~search'lnsiikrte cf MontTeal -. .

.: WCmc wrZling to rcquut pcmiiuiori t.rqodusc e mdptpuhlitbed in rhe BrifisI~Jod ofP7~cofogv so it may be hcded. as pst of a PhD bisof oac of ~~~.~IIWKS(Maiisa Fmdytus). Tho mannsl?n'pt tz~be rqxoduced is:

WCsequb wriücn pcz&sion fiam your o.£fice. to rq-xoduceit as pezt of th d~esis..As ihe rhcrCr submissiaa dcffdlirie Y dit csd ofJune, WC ask tbat you sfnd a Ictirr of pm?ission as soon u possible: P(case scnd.the infcr to thr Eollawjng addrrs:

&a.Quebn I32W IR7 CANADA

m.t!undrsWndhrg thetüw ell8cM tMmrr and mndtüans et0 6QWy

. . -...... - - 1 . . 7 10. w PO.QUW. Mnuu IQ~&I H~WIRT -. . . . + ._ - . ~~fi~~6!d1ant;b,ireo~~~twr:(5i&~en;aorc-~S, ...... - -. i - tJm& N*rdé de-UonbCPr at 6 ll-iht.Oie~de ~rnlnb.~ -.,. - - . ... . - ...___*._ - . . -.. . - . ..,.- '...... --...... - . - . -. .. . ) .-. .: .- ..* ..-...... - - .-- ..-...... t. . - 2. .m. ic . .- . -...... - ...... :c ....-. . - . A . - ...... - .- .-.- - . .------...... _ ...... T-. *.- ...... 14-sw.t-1596 16:52 WCMILLFIN PRESS LTD

..l. ! S. . 01. - - ..- . . I I .- , ...... *. .- , . . ... Stockton Press Houndmilis Basingrtoks Hampshire RG21 6XS UK Tel: 4-44 f011256 29242 Fîx: +44 (013 256 8 IO526

Macmillan Press:Umited grant permission for the reprodocfion of the attacheci material with the:following conditions:

(a) . the-eoun+]'of the material is ecknowledged by a full reference to the original artlcle. .

1 (b) permission tireproduce the materiel is obtained (as far as possible) from the c: auoior(s) of: the original article(s).

(c) no. eleçtronic:repruduction(of any sort) is pemitted. ( tiiut de recherches cliniques de Montréal CIinicaI Research lnstitute of Montreal

EIseYier Science B-V, Copy5@ and Penaissions Deptme~1t P.O. Box 521 IO00 AM --dam The Nethedan&

Fdytus ME, SchiUer PW, Shqiro M Welûowska G and Codene TJ (194%. The hÏghly selective hpioid antagonisr H-Tyr-Ticy [CHZ-WPhe-PheOH mP[y]) atcenuates morphine tolerance and dependence. Euz. J. Phmmcol. 286: IOj-10 8.

Ftmiyms ME znd Codene TJ (1996). Chnic inhibition of inùiicellulu ~a'; release or PKC activation significantly reduces the development of morphine dependence. En J. PhiucoZ. 3000: 173-181.

We reqaire writtm pemiission 60syom office to ïqroduce these as part of the thesu. -4s the thesis rubmission deadline is the end of Junc, we as.! you send a lemof permission as soon 3s possible. PIezse sznd the Imer ro the foilowing address:

Tel: (514) 987-5550 Fax (514) 987-5585

Thanr;-you for your cooperation

Terence J, Coderre, PhD 1 iO. avcnue des Pi75 CL'SS~. mon^& (Cwikc) HZW - 1 RF Tel6phoné: (514) 987-5500 T2ihpiec (57 4) 957-5678

AlfiliC Q L'Univeai. de MonkGar er .j IX5:lCDieu de ,Vonireal ELSEVIER SCIENCE 2

Elsetier Science - SL

Dr. TJ. Codeme Sara Bur~erh3mtrzut3 Pain Mechanisrns Loborcitory 105 KV Amsrsrdm Clinicul Research Instikife of Montreol The Secherlands 110 Pine Ave. West Montreal, Quebec H2W 1R7 Canada

Direct Line: (020) 4852 751 Amsterdam, 12 June 1996 Direct Fax: (020) 4852 722

Dear SirlMadam.

We herewith gront you and The National übraiy of Canada permission to reproduce the moteriol as mentioned in the enclosed copy of your letter in your thesisldissertation. provided that:

- full ocknowledgement is given to the original source of the publication - your work is not distributed commerciaIly

Shoukl your thesis be published commercially, please reapply for permission.

Youcs sincerely, ELSEVlER SCIENCE - NL Publishing Support & Services Deportment

w Mr. Jan van den Heuvel Rights & Permissions

Imprints: Ekerier Pergnmon North- Hullund Ercerpta Medica . . _. . Banken: Holiandxht Bank-Unie ...... IMAGE EVALUATION TEST TARGET (QA-3)

K lllllg L, 11111I.I LLY 1111-

APPLIED 1 IMAGE . lnc 1653 East Main Street .-. - Rochester. NY 14609 USA ,-- -- Phone: 7t W48~~ ------Fa: 716/208-5989

O 1993. Applied Image. Inc.. All Rlghts Resetved