Antinociceptive and nociceptive actions of opioids

Antinociceptive and Nociceptive Actions of Opioids

Michael H. Ossipov, Josephine Lai, Tamara King, Todd W. Vanderah,T. Philip Malan, Jr., Victor J. Hruby, Frank Porreca

Departments of Pharmacology, Anesthesiology and Chemistry, University of Arizona,Tucson, Arizona 85724

Received 15 April 2004; accepted 16 June 2004

ABSTRACT: Although the opioids are the princi-pal treatment options for moderate to severe pain, theiruse is also associated with the development of tolerance,defined as the progressive need for higher doses toachieve a constant analgesic effect. The mechanismswhich underlie this phenomenon remain unclear. Recentstudies revealed that cholecystokinin (CCK) is upregu-lated in the rostral ventromedial medulla (RVM) duringpersistent opioid exposure. CCK is both antiopioid and

pronociceptive, and activates descending pain facilita-tion mechanisms from the RVM enhancing nociceptivetransmission at the spinal cord and promoting hyperal-gesia. The neuroplastic changes elicited by opioid expo-sure reflect adaptive changes to promote increased paintransmission and consequent diminished antinociception(i.e., tolerance). © 2004 Wiley Periodicals, Inc. J Neurobiol 61:

126–148, 2004

Keywords: antinociception; nociception; opioids

ANTINOCICEPTIVE ACTIONSOF OPIOIDS

Opioid Receptors

The pain-relieving properties of the opioids, of whichmorphine is considered the prototype, have been ex-tensively studied and are well known. The existenceof endogenous receptors for opioids was reportedalmost simultaneously by three different groups in1973 (Pert and Snyder, 1973a; Simon, 1973; Tere-nius, 1973). The discovery of receptors selective forthe opioids, suggesting the existence of endogenouspain-relieving substances, generated great interest inunderstanding endogenous pain mechanisms. More-over, this interest extended into the study of the bio-logical mechanisms of opiate dependence and with-drawal. The probability of the existence of multiplesubtypes of the opioid receptor was first proposed

through purely pharmacologic studies reported byMartin and colleagues (Martin et al., 1976). It wasfound that vastly different behavioral syndromes wereelicited in spinalized dogs by morphine, ketocyclazo-cine and SKF-10,047 leading to the nomenclature ofmu, kappa, and sigma opiate receptors, respectively(McClane and Martin, 1967; Martin et al., 1976; Mar-tin, 1983). The observation of differential bindingaffinities for [Met5]enkephalin and �-endorphinagainst [3H][Leu5]enkephalin and [3H]naloxone inguinea pig brain and differential agonistic activityprofiles of [Leu5]enkephalin, [Met5]enkephalin andmorphine in the isolated mouse vas deferens andguinea pig ileum strongly supported the existence ofan additional receptor subtype, termed the delta (�)-opioid receptor (Lord et al., 1977; Waterfield et al.,1978, 1979). Over time, the existence of additionalopioid receptor subtypes has been hypothesized toinclude subtypes of these proposed receptors and asuggested receptor for �-endorphin, termed the epsi-lon receptor (Schulz et al., 1981). The advent ofmolecular cloning of the opioid receptors, however,led to the clear identification of only three subtypes ofthe opioid receptors (for review, see Massotte and

Correspondence to: F. Porreca ([email protected]).© 2004 Wiley Periodicals, Inc.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/neu.20091

126

Kieffer, 1998; Ossipov et al., 2004). The relationshipof the cloned opioid receptors to those identified phar-macologically remain to be elucidated (for review, seeZaki et al., 1996). Although different opioid receptorsubtypes have not been cloned, it is possible, forexample, that posttranscriptional events may occur toproduce pharmacologically distinct profiles (Zaki etal., 1996). Because the opioid receptors are widelydistributed throughout the entire nervous system, toinclude the peripheral terminations of sensory fibers,opioids may exert their antinociceptive activitythrough numerous mechanisms.

Opioid Receptors Are Prominent inPain Pathways

Spinal Sites of Action. The distribution and anatom-ical localization of the opioid receptors have beenextensively explored throughout the preceding de-cades through a variety of means, ranging from auto-radiography, radioligand binding, in situ hybridizationof mRNA for the receptors and immunohistochemis-try. The distributions of message coding for the �(Delfs et al., 1994; Mansour et al., 1995; Minami etal., 1994), � (Mansour et al., 1994a), and � (Minamiet al., 1993a, 1993b; DePaoli et al., 1994; Mansour etal., 1994b, 1995) opioid receptors have extensivelyreviewed (Satoh and Minami, 1995). Autoradio-graphic methods demonstrated that opioid receptorsare concentrated in the outer laminae of the dorsalhorns of the spinal cord (Besse et al., 1990a, 1991).Studies that employed tritiated selective agonists forthe opioid receptors indicated that the �-opioid recep-tor is highly concentrated in the outer laminae of thespinal dorsal horns, whereas the �-opioid receptor isdiffusely distributed throughout the dorsal horn(Quirion et al., 1983; Quirion, 1984). The �-opioidreceptor is believed to be concentrated in the outerlaminae of the dorsal horns of the lumbosacral cord,and is closely associated with nociceptive inputs fromthe viscera (Quirion et al., 1983; Quirion, 1984).Immunolabeling techniques provided more detailedinsights into the localization of the opioid receptorsand allowed the ultrastructural localization of the re-ceptors as well as the covisualization of two or morereceptor populations on the same neurons. All threeopioid receptors were found to be expressed primarilyin nociceptive C- and A� fibers of the dorsal rootganglia (DRG) cells as measured by immunohisto-chemistry (Dado et al., 1993; Arvidsson et al., 1995;Ji et al., 1995). Spinal morphine is believed to act inlarge part through opioid receptors residing on thecentral terminals of these primary afferent C-fibers(Yaksh and Noueihed, 1985; Lombard et al., 1995;

Mansour et al., 1995). Studies performed with patch-clamp techniques on isolated nociceptors found thatactivation of opioid receptors predictably inhibitedCa�� channels of small-diameter nociceptors and notof large-diameter cells, and suggest that receptor ac-tivation selectively inhibits the activity of C-fibers(Taddese et al., 1995). Autoradiographic inspection ofspinal cord slices after dorsal root rhizotomy from T13

to S2 suggested that the great majority of spinal opioid� (60%) and � (70%) receptors probably reside on thecentral terminals of afferent neurons (Besse et al.,1990a, 1991). The remaining receptors are believed toreside on either interneurons or on cell bodies ofsecond-order neurons that transmit nociceptive inputsto supraspinal sites that process nociceptive signals.Such studies also indicated that there is a greaterconcentration of �-opioid receptors in the outer lam-inae of the dorsal horn relative to the deeper layers(Dado et al., 1993; Arvidsson et al., 1995; Lai et al.,1996). This distribution is consistent with the hypoth-esis that � opioid agonists suppress the transmissionof pain signals from the primary sensory afferents thatterminate in laminae I and II onto projection neuronsthat form the spinothalamic tract. In contrast to thedistribution of the receptor protein, the levels ofmRNA for the �-opioid receptor were low to moder-ate throughout laminae I–VI, which would support thesuggestion that the preponderance of the �-opioidreceptors arise from dorsal root ganglion (DRG) neu-rons and are transported to the terminals of primaryafferents (Besse et al., 1990, 1991). Presumably,� receptors synthesized in the DRG are transported tothe central terminal of the primary afferent. The pre-synaptic � receptors are contained on the nociceptiveC-fibers, which terminate in laminae I and II. Impor-tantly, this distribution is consistent with the termina-tion fields of the unmyelinated C-fiber nociceptors,and not of the large-diameter, myelinated � fibers thatdo not transmit noxious inputs (Mansour et al., 1995).

Consistent with the presence of opioid receptors inthe spinal cord, spinally administered opiates exertrobust antinociceptive effects in animal models ofpain and in clinical practice as well. Morphine hasproduced dose-dependent, naloxone reversible antino-ciception after direct spinal administration in the rat,suggesting a modulation of nociceptive input or pro-cessing at the spinal level (Yaksh and Rudy, 1976,1977a). The direct spinal administration of [Met5]-enkephalin and related analogs also produced dose-dependent antinociception to noxious thermal stimulithat was reversed by systemic naloxone (Yaksh et al.,1977a). Advokat and colleagues reported that transec-tion of the spinal cord caused a loss in potency ofsystemic morphine against thermally evoked spinal

Actions of Opioids 127

nociceptive responses, whereas that of spinally in-jected morphine remained unchanged (Advokat andBurton, 1987). These observations supported the pre-vailing hypothesis that opioids may act directly atspinal sites to modulate nociceptive spinal reflexesand nociceptive inputs. Systemically administeredmorphine produced an attenuation of spontaneous andnoxious-evoked neuronal activity of lamina V inter-neurons (Le Bars et al., 1975). Moreover, systemicmorphine blocked the activity of second-order neu-rons in response to electrical stimulation of � andC-fibers of the sural nerve of decerebrate cats (Jurnaand Grossman, 1976). The iontophoretic applicationof morphine into the outer laminae of the dorsal hornsof the spinal cord has attenuated the responses ofdorsal horn units to noxious stimuli, even when suchneurons were detected within lamina V of the dorsalhorn (Duggan et al., 1976). These early studies pro-vided a clear demonstration that opioids may actspinally to alleviate pain. The spinal administration ofmorphine has now become routine medical practice inthe treatment of pain.

Supraspinal Sites of Action

Nociceptive signals entering at the level of the spinalcord are regulated not only by intrinsic interneuronsbut are also modulated by descending inhibitory pro-jections from supraspinal sites that are activated byopioid receptors. Autoradiographic studies have dem-onstrated that there are significant levels of opioidreceptor mRNA in many cortical, diencephalic, andbrainstem regions in addition to spinal loci (Quirion etal., 1983; Quirion, 1984). Mansour and colleagueshad performed exhaustive surveys of the brain utiliz-ing [3H]-labeled ligands for the �, �, and �-opioidreceptors and in situ hybridization methods formRNA for the receptors (Mansour et al., 1987, 1994a,1995). Regions that were shown to express the opioidreceptors include the frontal cortex, nucleus accum-bens, hippocampus, thalamus, and hypothalamus(Quirion et al., 1983; Quirion, 1984; Mansour et al.,1987, 1994b, 1985; Schmidt et al., 1994; Svingos etal., 1996). Two regions prominent in opioid-mediatedantinociception, the periaqueductal gray (PAG) andthe rostral ventromedial medulla (RVM) were identi-fied as expressing opioid receptors. The PAG wasfound to be rich in �-opioid receptors, whereas levelsof �- or �-opioid receptors were low or undetectable(Mansour et al., 1987). The RVM, which is defined asthe region of the medulla including the nucleus raphemagnus, the nucleus gigantocellularis pars alpha, andsurrounding reticular neurons ventral to the nucleusgigantocellularis and extending between the caudal

facial nucleus and the inferior olivary complex, wasfound to express all three subtypes of the opioidreceptors (Fields and Heinricher, 1985; Mansour etal., 1987). Later studies employing in situ hybridiza-tion for mRNA confirmed this differential distrubu-tion of the opioiod receptor subtypes within the PAGand RVM. These regions demonstrated extensive la-beling for the �-opioid receptor and light labeling forthe � or �- receptors (Mansour et al., 1995).

In addition, opioid receptors have also been iden-tified in the locus coeruleus, which also plays a role inmodulation of nociceptive inputs (Van Bockstaele etal., 1996). Many of the neurons that labeled for the�-opioid receptor also labeled tyrosine hydroxylase,and these were usually postsynaptic to unlabeled axonterminals that had characteristics of excitatory syn-apses (Van Bockstaele et al., 1996). These resultssuggest an important postsynaptic role for � opioidreceptor regulation of excitatory responses to cate-cholamine-containing neurons in the locus coeruleus.A critical consideration is that the locus coeruleus isthe source of the majority of the noradrenergic inner-vation of the brain, and it also expresses high levels ofmRNA for the �-opioid receptor. The noradrenergicneurons have terminals that innervate many otherbrain regions. These terminals may, in turn, have� opioid receptors on their surface, a fact that mayrelate to the lack of correlation between the distribu-tion of receptors and their mRNAs.

There is considerable evidence to indicate that theseregions are important for the antinociceptive effects me-diated by opioids. Importantly, reciprocal connectionsbetween the PAG and the RVM have been identified(Basbaum et al., 1978; Basbaum and Fields, 1978;Fields and Anderson, 1978). Opioids like morphinecause an activation of a population of cells in the PAGthat, in turn, excites neurons of the RVM (Fields andBasbaum, 1978; Basbaum and Fields, 1984; Fields et al.,1988). In 1976, Yaksh, Yeung, and Rudy conducted anextensive sterotactic survey of the brain by microinject-ing morphine into 403 sites and measuring nociceptiveresponses to noxious pinch (Yaksh et al., 1976). Theyfound that the ventrolateral aspect of the PAG mediateda robust, naloxone-reversible antinociceptive effect, inagreement with previous suggestions (Jacquet and La-jtha, 1973). Further exploration revealed that the PAGsites that were responsive to morphine also producedantinociception in response to electrical stimulation(Yeung et al., 1977). The microinjection of opioids intothe cerebral ventricles has been shown to produce dose-dependent antinociception in several species, includingmice and rats (Porreca et al., 1984; Dauge et al., 1987;Erspamer et al., 1989; Jiang et al., 1990; Miaskowski etal., 1991). A considerable amount of evidence exists to

128 Ossipov et al.

establish that opioid-induced antinociception is mediatedin part by descending pathways that originate from thePAG (Yaksh et al., 1976; Lewis and Gebhart, 1977a;Yaksh and Rudy, 1978). The microinjection or electricalstimulation of the ventrolateral aspect of the PAG hasproduced robust antinociceptive effects in rats (Lewisand Gebhart, 1977a, 1977b). Moreover, electrophysi-ologic studies demonstrated that the microinjection ofmorphine into the PAG attenuated the activity of pro-jection neurons in the dorsal horn in response to periph-eral nociceptive stimuli (Bennett and Mayer, 1979).Themicroinjection of morphine or of the �-opioid ago-nists [D-Pen2,D-Pen5]enkephalin (DPDPE) or [D-Ala2,Glu4]deltorphin into the medullary reticular formationblocked behavioral responses to noxious thermal stimuli,although the �-opioid agonists were inactive when ad-ministered into the PAG (Jensen and Yaksh, 1986a,1986b; Ossipov et al., 1995a). In addition to the PAG,the serotonergic nucleus raphe magnus (NRM) has beenshown to communicate with the PAG and have seroto-nergic projections to the spinal cord, and thus mayfunction as a relay for descending antinociceptive infor-mation arising from the PAG (Conrad and Pfaff, 1976a,1976b; Basbaum et al., 1977).

The RVM is recognized as a critical region withrespect to nociceptive processing and modulation, re-ceiving inputs from the spinal dorsal horn and fromrostral sites as well (Fields et al., 1983; Fields andHeinricher, 1985; Fields and Basbaum, 1999). Elec-trophysiologic studies where the responses of RVMneurons to noxious thermal stimulation have identi-fied the existence of “on”-cells and “off”-cells (Fields,1992; Fields and Basbaum, 1999; Heinricher et al.,2003). The off-cells are tonically active and pause infiring immediately before the animal withdraws fromthe noxious thermal stimulus, whereas the on-cellsaccelerate firing immediately before the nociceptivereflex occurs. An additional class, the “neutral” cellswere initially characterized by the absence of re-sponse to noxious thermal stimulation. It is now gen-erally understood that the activity of the off-cellscorrelate with inhibition of nociceptive input and no-cifensive responses, and these neurons may be thesource of descending inhibition of nociceptive inputs(Fields, 1992; Fields and Basbaum, 1999; Heinricheret al., 2003). In contrast, the response characteristicsof the on-cells suggest that these neurons are thesource of descending facilitation of nociception(Fields, 1992; Fields and Basbaum, 1999; McNally,1999; Heinricher et al., 2003). Accordingly, manipu-lations that facilitate responses to nociceptive stimulialso increase on-cell activity (Heinricher and Roy-chowdhury, 1997; Fields and Basbaum, 1999; Fields,2000; Heinricher et al., 2003). For example, pro-

longed delivery of a noxious thermal stimulus pro-duced increased on-cell along with a facilitation ofnociceptive reflexes (Morgan and Fields, 1994).Moreover, inactivation of RVM neuronal activitywith lidocaine blocked the facilitated withdrawal re-sponse (Morgan and Fields, 1994). It is now generallyaccepted that a spino-bulbo-spinal loop may be im-portant to the development and maintenance of exag-gerated pain behaviors produced by noxious (i.e.,hyperalgesia) and nonnoxious (i.e., allodynia) periph-eral stimuli (Urban and Gebhart, 1999; Ossipov et al.,2001; Porreca et al., 2002; Heinricher et al., 2003).Morphine microinjection or electrical stimulation inthe NRM has produced naloxone-sensitive antinoci-ception (Oliveras et al., 1977; Dickenson et al., 1979).Application of morphine or glutamate into the NRMor the nucleus gigantocellularis par alpha (NGCpA)has also produced antinociception in rats by acti-vating spinopetal mechanisms (Kiefel et al., 1993;McGowan and Hammond, 1993; Rossi et al., 1993).Accordingly, spinally administered 5-HT antagonistshave been shown to attenuate antinociception pro-duced by supraspinal or systemic morphine (Proudfitand Hammond, 1981). Studies employing selectiveantibodies for the �- and �-opioid receptors and ret-rograde tracing methods demonstrated the existenceof opioid-expressing neurons that project from thePAG or RVM to the spinal cord to provide a descend-ing inhibition of nociceptive inputs (Kalyuzhny et al.,1996). It was also proposed that opioids exert indirecteffects on PAG-RVM projection neurons, and haveboth direct and indirect effects on bulbospinal neurons(Kalyuzhny et al., 1996). Centrally administered mor-phine has also been demonstrated to activate descend-ing noradrenergic inhibitory pathways. The A5 (locuscoeruleus), A6, A7, and the catecholamine-containingnuclei in the vicinity of the lateral reticular nucleus(LRN) are major sources of noradrenergic innervationof the spinal cord, and have been implicated in thedescending inhibitory modulation of nociception (Ge-bhart and Ossipov, 1986; Jones and Gebhart, 1986a,1986b; Yeomans et al., 1992; Clark and Proudfit,1993).

SYNERGISTIC ACTIONS AT SPINAL/SUPRASPINAL SITES

An important aspect of morphine-mediated antinoci-ception is that antinociceptive efficacy when givensystemically is greater than that which would be pre-dicted based solely on its in vitro activity in isolatedtissues. The efficacy of morphine is believed to be a

Actions of Opioids 129

result of a synergistic interaction between its activityat spinal and supraspinal sites. This spinal/supraspinalsynergy was initially demonstrated by Yeung andRudy (1980a) when it was found that naloxone giveneither spinally into the cerebral ventricles did notattenuate the antinociceptive actions of systemic mor-phine in a manner consistent with an additive effectbetween these sites (Yeung and Rudy, 1980a). It wasfound that a 1:1 fixed ratio of morphine administeredintrathecally (i.th.) and i.c.v. produced an approxi-mate 30-fold increase in potency when evaluated inthe tail-flick test and a 45-fold increase in the hot-plate test compared to i.c.v. morphine alone (Yeungand Rudy, 1980a). Isobolographic analyses employ-ing several dose ratios demonstrated a hyperbolicfunction with a strong degree of curvature, and it wasconcluded that potentiation would occur at all possi-ble combinations of spinal and supraspinal levels ofmorphine after systemic injection. This observation ofa multiplicative interaction between spinal and su-praspinal morphine has also been observed in mice(Roerig and Fujimoto, 1988), and been extendedto include the highly selective �-opioid agonistDAMGO (Roerig and Fujimoto, 1989). Further, evi-dence that the synergistic antinociceptive activity ofmorphine is the result of a site–site interaction wasconfirmed when synergy was determined based onspinal and supraspinal morphine content, rather thandoses administered, in rats (Miyamoto et al., 1991).Repeated exposure to morphine produced a loss ofspinal/supraspinal synergy. Rats received systemicmorphine and challenged with PAG and spinal mor-phine. Those with morphine “tolerance” showed aloss of the synergistic interaction between spinal andPAG morphine (Siuciak and Advokat, 1989). A sim-ilar observation was made with regard to neuropathicpain. Rats with ligation of the L5 and L6 spinal nerves(SNL) demonstrate abnormal, enhanced pain behav-iors, some of which are resistant to opioids. Specifi-cally, spinal morphine fails to attenuate tactile hyper-sensitivity in rats with SNL, and in these animals thespinal/supraspinal synergy was absent and the po-tency of systemic morphine was diminished (Bian etal., 1999). Restoration of the spinal site of morphineactivity with an NMDA antagonist restored supraspi-nal-spinal synergy and increased the potency of sys-temically given morphine in animals with nerve injury(Bian et al., 1999). Critically, these studies showedthat systemic opiates were active in neuropathic con-ditions, a finding that has now been confirmed inhumans (Rowbotham, 1995; Dworkin et al., 2003;Rowbotham et al., 2003).

Nociceptive Actions of Opioids

Antinociceptive and Analgesic Tolerance. The opi-oids continue to hold a prominent position in thetreatment of both acute and chronic pain states. Mor-phine, given through a variety of drug delivery sys-tems or fentanyl, typically given in a transdermalsystem, represent the most efficacious and aggressivemethods to treat severe pain conditions. A significanthindrance, however, with regard to the prolonged useof such opioids is the development of tolerance totheir analgesic effects. Tolerance is defined as a de-crease in analgesic activity of a drug after a previousexposure to the same or a similar drug (Way et al.,1969; Cox, 1990; Foley, 1993, 1995). The repeateddaily systemic injections of morphine to mice or ratsproduced significant rightward shifts in the antinoci-ceptive effect of morphine challenge in two tests ofnociception, the hot-plate test, and acetic acid-inducedwrithing (Fernandes et al., 1977a, 1977b). Repeatedsystemic or i.th. injections of morphine also produceda rightward shift in the dose–response curves for i.th.morphine in the hot-plate and tail-flick tests (Yaksh etal., 1977b). Clinically, antinociceptive tolerance man-ifests as a diminished or lost pain relief of a givenopioid dose administered repeatedly or with continu-ous administration of an opioid over some period oftime. Opioid analgesic tolerance is well recognizedexperimentally and clinically, and can occur over aperiod of days to weeks (Way et al., 1969; Foley,1993, 1995). Clinically, the need for increasing dosesof opioids in cases of chronic pain is well documentedand usually presented as a major obstacle to providingadequate pain relief over a long period of time (Foley,1993, 1995; Cherney and Portenoy, 1999). By way ofillustration, a review of the clinical experience of over700 patients that received spinal morphine over anaverage of 124 days revealed that analgesic toleranceto spinal opioid use developed to different degreesamong patients, and appeared to be related to the typeof pain and differences in pharmacokinetics amongpatients (Arner et al., 1988). Despite much intensiveresearch, however, the mechanisms that underlie thedevelopment of tolerance to the analgesic effects ofopioids remain largely unknown. There are some clin-ical reports that suggest that tolerance to morphinemay not always develop. A 10-year prospective studythat included 2118 patients receiving palliative care,56% of whom received morphine, was performed tovalidate the World Health Organization (WHO)guidelines (Zech et al., 1995). It was reported that aneed for increased doses of morphine occurred inone-half the patients, whereas the remainder main-tained a stable or decreasing dosing requirement

130 Ossipov et al.

(Zech et al., 1995). However, these patients also re-ceived, at various times, ancillary treatments that in-cluded antidepressants, anticonvulsants, antihista-mines, antiemetics, or corticosteroids, which coulddiminish or prevent tolerance development (Zech etal., 1995). Patients with inflamed joints were found tohave elevated levels of endogenous opioid peptides inthe synovial fluid (Stein et al., 1996). Nevertheless,the analgesic effect of intraarticular morphine in thispopulation, leading to the assertion that opioid pep-tides expressed in inflamed tissue do not producetolerance to peripheral morphine (Stein et al., 1996).In a study of gastrointestinal (GI) transit in mice, itwas found that intestinal inflammation reversed toler-ance to morphine-induced inhibition of GI motility(Pol and Puig, 1997). However, the apparent reversalof tolerance may be related to the observations thatconditions of inflammation are associated with re-duced availability of CCK and enhanced activity ofmorphine (Stanfa et al., 1994; Ossipov et al., 1995a;Vanderah et al., 1996a). This interpretation is consis-tent with the fact that CCK antagonists reversed tol-erance to morphine-induced inhibition of GI motility(Pol and Puig, 1997).

Many studies have focused on changes occurringat the cellular level to gain an appreciation of themechanisms that drive the development of antinoci-ceptive tolerance (Sabbe and Yaksh, 1990; Childers,1991; Collin and Cesselin, 1991). Although alter-ations in subcellular processes undoubtedly contributeto changes in physiology that occur during prolongedexposure to opioids, the present level of understand-ing of such processes is insufficient to allow for thedirect correlation of intracellular changes to thoseoccurring at the level of the neuronal circuits mediat-ing antinociception or analgesia. Further complicatingpossible translation of intracellular changes that mayunderlie the development of tolerance is the puzzlingobservation that many, seemingly unrelated classes ofsubstances, including antagonists of classical and pep-tidergic neurotransmitters and enzyme inhibitors havethe ability to “block” antinociceptive tolerance.Among the substances reported to “block” opioidantinociceptive tolerance are CGRP antagonists (Me-nard et al., 1996; Powell et al., 2000), nitric oxidesynthase inhibitors (Powell et al., 1999), calciumchannel blockers (Aley and Levine, 1997), cyclooxy-genase inhibitors (Powell et al., 1999), protein kinaseC inhibitors (Mao et al., 1995c), competitive andnoncompetitive antagonists of the N-methyl, D-aspar-tate (NMDA) receptor, AMPA antagonists (Kest etal., 1997), superoxide dismutase mimics (Dr. D.Salvemini, personal communication), dynorphin anti-serum (Vanderah et al., 2000), and CCK antagonists

(Dourish et al., 1988; Xu et al., 1992). Blockade ofopioid tolerance by antagonists of NMDA receptorshas been especially well studied, and proposed mech-anisms have focused on the likely colocalization ofNMDA and opioid receptors (Trujillo and Akil, 1991;Mao et al., 1995a, 1996; Lutfy et al., 1996; Manninget al., 1996). It has been suggested that prolongedexposure to opioids causes an NMDA-receptor linkedtranslocation of PKC from cytosol to membranewhere it becomes activated (Mao et al., 1995a; Mao etal., 1995b; Mayer et al., 1995a,b). Blockade of PKCtranslocation with the GM1 ganglioside or blockadeof NMDA receptors have prevented the developmentof tolerance to morphine (Mao et al., 1995a, 1995b;Mayer et al., 1995b). A number of intracellularchanges that may occur after persistent opioid expo-sure, including changes in second-messenger systemshave been extensively reviewed recently (Williams etal., 2001; Kieffer and Evans, 2002).

As so many systems can be invoked to modulateopioid antinociceptive tolerance, it becomes difficultto implicate a common cellular mechanism for theactions of all these substances. Efforts to reveal phys-iological mechanisms that might underlie the phe-nomenon of opioid antinociceptive tolerance thereforeare justified. There is emerging evidence that peptidicneuromodulators such as CCK or dynorphin may actas an endogenous physiologic antagonists of endoge-nous or exogenous opioid activity. More precisely,these substances may be classified as “pronocicep-tive” and by enhancing nociception; these and othersubstances may drive the expression of antinocicep-tive tolerance to opioids. In this regard, it should berecognized that pain may be considered as a physio-logical antagonist of analgesia and increased states ofpain require increased levels of pain relieving opiate,resulting in “opiate tolerance” (Vanderah et al.,2001a).

Opioid-Induced Paradoxical Pain. An importantconcept that has been gaining considerable experi-mental validation is that exposure to opioids eitheracutely or for prolonged periods paradoxically elicitshyperalgesia and other signs of abnormal pain. Thisparadoxical pain, and the neurobiological adaptationsthat underlie this state, may play an important role inthe requirement for increased levels of opioids tomaintain a constant degrees of antinociception, that is,tolerance (Vanderah et al., 2000, 2001a, 2001b;Gardell et al., 2002).

Clinical Evidence for Opiate-Induced AbnormalPain. A number of clinical reports exist that showthat opioid administration in a clinical setting can

Actions of Opioids 131

elicit an unexpected paradoxic abnormally heightenedpain sensations. In a review of the clinical experienceof 750 patients receiving epidural morphine, and18 patients receiving spinal morphine, morphine-in-duced abnormal pain was documented (Arner et al.,1988). Many patients developed hyperesthesias (in-creased sensitivity to sensory stimuli such as lighttouch or brush) and allodynia (pain elicited by nor-mally innocuous sensory stimulation). Severe allo-dynia unrelated to the original pain complaint oc-curred in one patient after receiving an intrathecalinfusion of 30 mg of morphine (Arner et al., 1988). Inanother clinical report, a patient with an original paincomplaint from advanced epidermoid carcinoma ofthe right lung and thoracic pain developed spontane-ous pain, hyperesthesia, and allodynia of the legs witha dermatomal distribution of pain corresponding fromthe lumbosacral region after receiving continuoussubarachnoid infusion of morphine (De Conno et al.,1991). Hyperesthesias were also induced by long-term intrathecal morphine infusion in a cancer patientwhere the pain was originally controlled by 1 mg/dayof intrathecal morphine (Ali, 1986). Again, the abnor-mal pain was manifested in a different way than theoriginal pain complaint, and was described as a burn-ing sensation of the whole leg, rather than an inter-mittent shooting type of pain that was initially de-scribed (Ali, 1986). More recently, a report waspublished where a spinal infusion of sufentanil in apatient with neuropathic pain secondary to arachnoid-itis and laminectomy originally alleviated the pain butthen evoked hyperesthesias in the lower extremities(Devulder, 1997). This abnormal pain state was de-scribed as being qualitatively different from the orig-inal complaint and included the back, abdomen, andboth legs. Cancer patients that received high doses ofintrathecal morphine by bolus injections also reportedparadoxical intense pain within one-half hour of theinjections (Stillman et al., 1987). The possibility thatopioids may produce abnormal pain after even short-term exposures has been suggested as a plausiblereason as to why clinical results of studies on preemp-tive analgesia have been disappointing (Eisenach,2000). It is important to note, however, that althoughhigher doses of intraoperative fentanyl may be asso-ciated with lower pain thresholds than at lower doses,patients receiving intraoperative fentanyl still demon-strate higher pain thresholds than those receiving none(see Eisenach, 2000).

Opioid-Induced Abnormal Pain Is Confirmed in An-imal Studies. A considerable number of animal stud-ies have clearly demonstrated that opioid administra-tion may produce an abnormal, paradoxic pain state.

In some instances, opioid-induced abnormal pain hasbeen likened to the behavioral symptomology ob-served after peripheral nerve injury (Mao et al.,1992a, 1994; Mayer et al., 1995a, 1995b; Wegert etal., 1997; Ossipov et al., 1998 Vanderah, 2001; Ossi-pov et al., 2000b). Large doses of intrathecal mor-phine have caused paradoxical algesia and hyperes-thesias (Woolf, 1981). The spinal administration of abolus dose of 50 �g of morphine to rats elicited painbehavior that involved intermittent bouts of biting andscratching at the dermatomes corresponding to theinjection site along with aggressive and nocifensivebehaviors in response to light brushing of the flanks(Yaksh et al., 1986). Higher doses of i.th. morphine(90 to 150 �g) also provoked periodic bouts of spon-taneous agitation and nocifensive responses to lighttouch (Yaksh and Harty, 1988). Moreover, rats thatwere made tolerant to either systemic or spinal mor-phine demonstrated hyperreflexia and extreme sensi-tivity to handling upon the injection of either spinal orsystemic naloxone (Yaksh et al., 1977b). Subse-quently, studies employing either continuous infusionor repeated injections of morphine demonstratedenhanced pain. The repeated daily injection of spi-nal morphine produced enhanced responses of thetail or hindpaw to noxious thermal stimuli withineight days (Mao et al., 1994; Mayer et al., 1995a).Likewise, repeated daily systemic injections ofmorphine produced a loss of its antinociceptiveeffect that progressed to enhanced behavioral re-sponses to noxious radiant heat (Trujillo and Akil,1991, 1994).

An interesting phenomenon has been describedwhere a single systemic dose of opioid produced signsof hyperalgesia after the initial opioid-induced antino-ciceptive action has subsided, suggesting that the hy-peralgesic component is present after opioid admin-istration but masked by the analgesia observed in theinitial postopioid period (Larcher et al., 1998; Celerieret al., 1999, 2000). In one study, the systemic injec-tion of heroin produced antinociception in the tail-flick test and the antinociceptive effect of a subse-quent dose given immediately after the effect of theinitial dose was terminated produced a significantlylower effect (Larcher et al., 1998). Additionally, theinjection of naloxone after a single dose of heroinseemed to unmask a hyperalgesia, indicated by short-ened tail-flick latencies (Larcher et al., 1998). In aseparate study, the s.c. injection of heroin producedantinociception that was followed by decreasedthresholds to evoke paw-pressure induced vocaliza-tion, which was interpreted as opioid-induced allo-dynia (Laulin et al., 1998). The same group alsodemonstrated that a single injection of morphine pro-

132 Ossipov et al.

duced an initial antinociception followed by thermalhyperalgesia (Celerier et al., 1999). Furthermore, inthe same study, it was reported that the administrationof naloxone during the antinociceptive phase of mor-phine or fentanyl unmasked an NMDA-mediated hy-peralgesic effect (Celerier et al., 1999). Most recently,in an effort to determine if fentanyl elicits a sensiti-zation to pain, rats received four injections of fentanyl15 min apart and nociceptive thresholds to paw pres-sure were determined at selected time intervals afterinjection and daily afterwards (Celerier et al., 2000).Impressively, significant hyperalgesia persisted for upto 5 days after the fentanyl injections (Celerier et al.,2000). These studies indicate that even a short-termexposure to opioids may produce a rebound hyperal-gesisic effect, and this may manifest behaviorally astolerance.

Reports of “opioid-induced” hyperalgesia havebeen suggested to be the result of an unmasking of acompensatory neuronal hyperactivity in response tomorphine-induced inhibition of neuronal function(Gutstein, 1996). This hyperresponsiveness, or sensi-tization, becomes evident after the opioid is removedor occurs intermittently between injections such thatopioid-induced hyperalgesia might be interpreted as aresult of repeated episodes of opioid withdrawal, acondition identified as “miniwithdrawals” (Gutstein,1996). Studies of abnormal pain consequent to long-term opioid administration that depend on paradigmsof repeated injection are generally subject to thiscriticism (Trujillo and Akil, 1991; Mao et al., 1994;Mao et al., 1995a). Recent studies emerging from ourlaboratories have demonstrated that continuous expo-sure to opioids produced behavioral signs of exagger-ated pain and, importantly, that such pain occurredwhile the opioid was continuously present in the sys-tem (Vanderah et al., 2000, 2001b; Gardell et al.,2002). The continuous spinal infusion of [D-Ala2,N-Me-Phe4,Gly-ol5]enkephalin (DAMGO) deliveredthrough an osmotic minipump to rats produced an-tinociceptive tolerance to DAMGO or morphine, asdemonstrated by a reduction in their antinociceptiveeffect within 6 days, and by a rightward shift in themorphine dose–response curve against the tail-flicktest (Vanderah et al., 2000). Concurrently, these ani-mals expressed tactile allodynia and thermal hyperal-gesia, indicated by significant reductions in paw with-drawal responses to light tactile or noxious radiantheat applied to the hindpaws (Vanderah et al., 2000).Importantly, these behavioral signs of abnormal painwere present while DAMGO was still being infused(Vanderah et al., 2000). Similar effects were seenwhen prolonged exposure to morphine was producedthrough the subcutaneous administration of morphine

(Vanderah et al., 2001b). Morphine was continuouslyinfused through an osmotic minipump or releasedfrom a pair of subcutaneously implanted pellets con-taining free-base morphine. Within 7 days, the ratsdemonstrated reduced response thresholds to lighttactile or noxious radiant heat stimuli, indicating thepresence of tactile allodynia and thermal hyperalgesia(Vanderah et al., 2001b). As with spinal DAMGOinfusion, continuous exposure to systemic morphinealso produced a significant rightward shift in the in-trathecal or systemic morphine dose–response curves(Vanderah et al., 2001b). In both these studies, abnor-mal pain was present while the opioid was still beingadministered to the animals. The demonstration ofabnormal pain during the continuous delivery of opi-oids, by multiple routes and through minipumps andpellets, minimizes concerns that the sensory changeswere due to the development of states of “miniwith-drawals.”

It should be noted that some studies suggest thatthe repeated administration of opioids do not causehyperalgesia, but rather a potentiation of antinocicep-tive activity (Kayser and Guilbaud, 1985; Kayser etal., 1986; Neil et al., 1990; Gutstein et al., 1995).Initially based on drug discrimination studies, Col-paert and colleagues suggest that the phenomenon oftolerance to opioids simply does not exist, and insteadpropose a “Systems Theory” where tolerance does notdevelop to the primary physiologic action of opioids(Colpaert et al., 1976; Colpaert, 1978). They suggestinstead that chronic administration of opioids causesan adaptive hyperalgesia, resulting in an apparent lossof antinociceptive effect of morphine (Colpaert, 1995,1996, 2002). Conversely, a persistent noxious stimu-lus causes an adaptive hypoalgesia (Colpaert, 1995,1996, 2002). Consequently, overestimating the mor-phine dose required to just block chronic pain causesa mismatch between morphine-induced hyperalgesiais insufficiently offset by pain-induced hypoalgesia,and an apparent tolerance to morphine is observed(Colpaert, 1995, 1996, 2002). The proposed adaptivemechanisms that might underlie such adaptations,particularly pain induced hypoalgesia, have not yetbeen clarified.

Mechanisms of Opioid-Induced Paradoxical Pain.Abnormal pain is promoted by descending facilitationfrom the RVM: the descending facilitatory mecha-nism from the RVM (described above) may also sig-nificantly underlie the development of abnormal painstates subsequent to opioid exposure. It has beenreported that spontaneous activity of on-cells in-creases along with facilitated nocisponsive behaviorduring naloxone-precipitated withdrawal (Bederson et

Actions of Opioids 133

al., 1990; Kim et al., 1990). Moreover, these actionswere blocked by microinjection of lidocaine into theRVM (Kaplan and Fields, 1991). Despite these inves-tigations, the state of on-cell or off-cell firing duringsustained morphine administration, in the absence ofwithdrawal, has not been directly studied. Continuousexposure to morphine by s.c. pellet implantation orosmotic minipump has been shown to produce tactileallodynia and thermal hyperalgesia while the ratswere still receiving morphine (Vanderah et al.,2001a). The microinjection of lidocaine into the RVMproduced a reversible block of both tactile allodyniaand thermal hyperalgesia in morphine-tolerant rats(Vanderah et al., 2001b). Such abnormal pain devel-oped over a period of days, and did not reflect theacute activity of the opioid (Vanderah et al., 2001b).Moreover, the s.c. implantation of morphine pelletsproduced significant rightward shifts in the antinoci-ceptive dose–response curves for morphine given ei-ther i.th. or s.c. (Vanderah et al., 2001b). However, inthe presence of lidocaine in the RVM, these dose–response curves were shifted to the left such that theA50 values were not significantly different from thoseof nontolerant rats (Vanderah et al., 2001b). Finally,the ability of lidocaine microinjected into the RVM torestore the antinociceptive potency of i.th. morphinewas reversible and consistent with the duration ofaction of lidocaine (Vanderah et al., 2001b).

Similar results were obtained by performing bilat-eral lesions of the dorsolateral funiculus (DLF), fur-ther supporting this hypothesis. Spinopetal projec-tions from RVM neurons make up the majority offibers of the dorsolateral funiculus (DLF), and maycontain descending facilitatory fibers from this region(Fields and Heinricher, 1985). To further investigatethis mechanism, rats made tolerant to the antinocicep-tive action of morphine by s.c. pellets expressed be-havioral signs of abnormal pain (Vanderah et al.,2001b). Bilateral disruption of the DLF prevented thedevelopment of both tactile allodynia and thermalhyperalgesia resulting from sustained opioid deliverywithout affecting normal responses to acute noxiousor innocuous stimuli (Vanderah et al., 2001b). Fur-thermore, bilateral DLF lesions prior to morphinepellet implantation prevented the development andexpression of morphine antinociceptive tolerance asshown by a lack of rightward displacement of the i.th.dose–response curve compared to rats with sham DLFlesions (Vanderah et al., 2001b). Normal nocifensiveresponses and the antinociceptive action of morphinein rats implanted with placebo pellets was not affectedby DLF lesions, indicating that these changes werenot due to a disruption of normal sensory processing(Vanderah et al., 2001b). These observations are sim-

ilar to other situations where abnormal pain wascaused by different means. Tactile allodynia causedby spinal nerve ligation was abolished by DLF lesion(Ossipov et al., 2000a). Likewise, the electrical stim-ulation of the DLF produced clear excitation of neu-rons in the superficial laminae of the dorsal horn,demonstrating a clear descending facilitation throughthis pathway (McMahon and Wall, 1983, 1988).Taken together, these data support the hypothesis thatdescending facilitation from the RVM serves to pro-mote a facilitated pain state manifested as opioid“tolerance.”

As noted above, the on-cells of the RVM are animportant source of descending pain facilitatory pro-jections. Evidence exists to show that these cellsmight be activated by CCK, because the microinjec-tion of CCK8 into the RVM has enhanced nociceptiveinput and attenuated the morphine-induced reductionof on-cell responses to nociception (Heinricher andMcGaraughty, 1996). Behavioral signs of tactile allo-dynia and thermal hyperalgesia were produced byCCK8 in the RVM as well (Kovelowski et al., 2000).In contrast, the microinjection of the CCKB receptorantagonist, L365,260, into the RVM blocked bothtactile allodynia and thermal hyperalgesia in rats withL5/L6 SNL (Kovelowski et al., 2000). Lidocaine in theRVM also reversibly blocked these behavioral signsof neuropathic pain, presumably by blocking facilita-tion arising from the RVM (Pertovaara et al., 1996;Kovelowski et al., 2000). Electrical stimulation of theRVM at low intensities has facilitated dorsal hornneuronal activity and the spinal nociceptive tail-flickreflex, further demonstrating the existence of nocicep-tive facilitation arising from this region (Zhuo andGebhart, 1992, 1997).

Spinal dynorphin and opioid-induced paradoxicalpain: considerable evidence has demonstrated thatenhanced expression of spinal dynorphin is pronoci-ceptive and appears to promote facilitated pain states.States of chronic inflammation and peripheral nerveinjury that are accompanied by manifestations of ab-normal pain, including spontaneous pain, allodynia,and hyperalgesia, are also associated with elevatedspinal dynorphin content (Kajander et al., 1990;Draisci et al., 1991; Dubner and Ruda, 1992). Cryo-neurolytic disruption of the sciatic nerve has led toelevated spinal dynorphin content, and associatedpain behaviors were blocked by antisera to dynorphin(Wagner et al., 1993; Wagner and Deleo, 1996).Dynorphin-like immunoreactivity and prodynorphinmRNA levels were elevated in the spinal cord perfus-ate of polyarthritic rats (Pohl et al., 1997). A singlespinal injection dynorphin has produced long-lastingtactile allodynia in rats and mice (Vanderah et al.,

134 Ossipov et al.

1996b; Laughlin et al., 1997). Enhanced abnormalpain states that result from peripheral nerve injury areassociated with enhanced spinal dynorphin levels, andthose procedures or treatments that interfere with el-evated spinal dynorphin content or function also abol-ish injury-induced abnormal pain (Malan et al., 2000;Porreca et al., 2001; Wang et al., 2001; Burgess et al.,2002; Gardell et al., 2003).

Elevations in spinal dynorphin content are alsoseen in animals with prolonged, constant exposure toopioids either systemically or spinally (Vanderah etal., 2001a, 2001b; Gardell et al., 2002). Spinal infu-sion of the mu opioid agonist, DAMGO over 6—7 days was shown to elicit tactile allodynia and ther-mal hyperalgesia while the opioid infusion was con-tinuing (Vanderah et al., 2000). This treatment alsoproduced elevated dynorphin content in the lumbarcord as well as immunoreactivity for prodynorphin(Vanderah et al., 2000). The spinal injection of anti-serum to dynorphin blocked tactile allodynia and ther-mal hyperalgesia in the DAMGO-treated rats, but didnot elicit any changes in nontolerant rats. More im-portantly, antiserum to dynorphin unmasked the an-tinociceptive action of the DAMGO that was stillinfused (Vanderah et al., 2000) and blocked the right-ward displacement of the dose–effect curve for spinalmorphine in DAMGO infused rats, indicating a block-ade of antinociceptive tolerance (Vanderah et al.,2000). Antiserum to dynorphin did not alter the an-tinociceptive activity and potency of spinal morphinein vehicle-infused rats (Vanderah et al., 2000). Bilat-eral lesions of the DLF, which were shown to blockabnormal pain and tolerance to the antinociceptiveeffect of morphine, also prevented the upregulation ofspinal dynorphin (Gardell et al., 2003). Thus, manip-ulations that block opioid-induced pain, in this casedue to spinal infusion of opioid, also block the behav-ioral manifestation of antinociceptive tolerance. Thedata show that sustained opioid administration leadsto elevated spinal dynorphin content, which in turn,promotes an abnormal pain state. This enhanced painstates increases the requirement for opioid dose pro-duce a comparable antinociceptive effect seen in an-imals without enhanced nociception, resulting in anapparent manifestation of antinociceptive tolerance. Itshould be emphasized that such pain occurred whilethe opioid was continually delivered to the spinalcord, arguing against opioid withdrawal as an expla-nation of altered sensory level.

Elevated spinal dynorphin levels are the result ofdescending facilitation: as noted above, systemic ad-ministration of opioids elicit an increased expressionof spinal dynorphin, and this may be the result oftonic descending facilitation arising in brainstem

sites. The precise mechanisms through which in-creased spinal dynorphin expression promotes pain,and consequently, the manifestation of opioid toler-ance, remains to be elucidated. However, there isevidence that increased spinal dynorphin promotes thefurther release of excitatory transmitters from primaryafferent neurons, in this way provoking a positivefeedback loop that amplifies further sensory input.Microdialysis studies have demonstrated localized,dose-dependent release of glutamate and aspartateelicited by exogenous dynorphin in the hippocampusand spinal cord (Faden, 1992; Skilling et al., 1992).More recently, the capsaicin-stimulated release of cal-citonin gene-related peptide (CGRP) was potentiatedby dynorphin A(2–13), a nonopioid fragment, in spinalcord slices in vitro (Claude et al., 1999; Gardell et al.,2002, 2003). In these studies, CGRP was employed asa marker for release of excitatory transmitter fromprimary afferent neurons (Claude et al., 1999; Gardellet al., 2002, 2003).These observations are consistentwith previous reports of dynorphin facilitation of cap-saicin-evoked substance P release from trigeminalnuclear slices, an effect blocked by MK-801 but notby opioid antagonists (Arcaya et al., 1999). Mostrecently, it was demonstrated that persistent exposureto morphine pellets implanted subcutaneously pro-duced enhanced capsaicin-evoked release of CGRPfrom spinal tissue (Gardell et al., 2002, 2003). Thisenhanced evoked release was blocked by the additionof antiserum to dynorphin in the perfusion medium.Moreover, the disruption of descending facilitationfrom suprsapinal sites by selective ablation of RVMneurons that express the mu-opioid receptor or bysurgical lesions of the DLF prevented opioid-inducedabnormal pain, spinal dynorphin upregulation, andenhanced capsaicin-evoked release of CGRP (Gardellet al. 2002, 2003). Finally, enhanced capsaicin-evoked release of CGRP was also blocked by theNMDA antagonist MK-801 (Gardell et al., 2002,2003).

Enhanced pain from increased spinal excitatoryamino acid activity: the loss of antinociceptive activ-ity after sustained spinal opioid administration is qual-itatively similar to the diminished effect of spinalopioids in animal models of neuropathic pain. Bothstates show diminished opioid analgesic potency andefficacy, and both states are associated with abnormalpain including thermal hyperalgesia and tactile allo-dynia, suggesting the possibility of common mecha-nisms in the postnerve injury state and in spinal �opioid tolerance (Mao et al., 1995a; Mao et al., 1995b;Mayer et al., 1995a; Wegert et al., 1997; Ossipov etal., 1998, 2000b). A prominent similarity betweenopioid tolerance and nerve injury is that abnormal

Actions of Opioids 135

pain is likely the result of central sensitization conse-quent to the release of glutamate acting upon theNMDA receptor complex. In support of this concept,it has been found that the i.th. injection of NMDAantagonists reversed established thermal hyperalgesiaelicited by peripheral nerve injury (Mao et al., 1992a,1993, 1995a; Wegert et al., 1997; Bian et al., 1999).Similarly, the loss of antinociceptive potency of mor-phine to suppress the tail-flick response in animalswith an nerve injury to lumbar spinal nerves wasrestored by i.th. MK-801 (Wegert et al., 1997). Sim-ilar to peripheral nerve injury, blockade of NMDAreceptors has also prevented opioid tolerance or ab-normal pain due to prolonged opioid exposure (Tru-jillo and Akil, 1991, 1994; Tiseo and Inturrisi, 1993;Mao et al., 1994, 1996, 1998; Tiseo et al., 1994).Further, once daily spinal injection of morphine tonormal rats produced antinociceptive tolerance alongwith thermal hyperalgesia of the hindpaws (Mao etal., 1994), and both of these effects were prevented bycoadministration of MK801. Continuous i.th. infusionof morphine also reliably produced tolerance to itsantinociceptive effect, and the development of toler-ance was prevented by the coinfusion of the NMDAantagonists MK801 or dextromethorphan (Manning etal., 1996). Competitive and noncompetitive blockersof the NMDA receptor both blocked antinociceptivetolerance to morphine as determined in the formalin-induced flinch model (Lutfy et al., 1996). Thus, theNMDA receptor complex modulates hyperalgesia as-sociated with neuropathic pain states as well as opioidtolerance. In further support of the role of NMDAreceptors, Simonnet and colleagues (Larcher et al.,1998; Celerier et al., 2000) also demonstrated that thelong-term allodynia occurring after a single dose ofheroin or fentanyl was prevented by MK-801. Collec-tively, these studies suggest that exposure to opioidsresult in an enhanced sensitivity of the spinal cord tonociceptive inputs, and that this spinal sensitization ismediated, at least in part, through NMDA receptors.

Blocking Opioid-Induced EnhancedAbnormal Pain Restores MorphineEfficacy

The condition of elevated nociceptive input causingincreases in the requirement for opioid dose in ordermaintain analgesic activity may result in expression ofantinociceptive tolerance. The s.c. presence of mor-phine pellets over 7 days has resulted decreased pawwithdrawal latencies to noxious radiant heat (Van-derah et al., 2001b). Correspondingly, these animalsalso demonstrated a significant shift to the right of thei.th. morphine antinociceptive dose–response curve

when compared to that of placebo-implanted rats(Vanderah et al., 2001b). However, when the intensityof the noxious radiant heat source is adjusted so thatthe paw withdrawal latencies of the placebo-im-planted and morphine-implanted rats are not signifi-cantly different, then the antinociceptive dose–re-sponse curves for i.th. morphine between the twogroups of animals do not differ statistically (Porrecaand Vanderah, unpublished observations). These ob-servations support the hypothesis that opioid antino-ciceptive “tolerance” may be construed as arising as aconsequence of increased sensitivity to nociceptiveinput after exposure to opioids. This view is alsosupported by more detailed analysis of the substancesthat have been demonstrated to “block” opioid toler-ance. Some of these are considered individuallybelow.

NMDA Antagonists. It has long been appreciatedthat activation of the N-methyl, D-aspartate (NMDA)receptor by glutamate results in the “sensitization” ofneurons (Wilcox, 1991; Ma and Woolf, 1995;Baranauskas and Nistri, 1998). NMDA receptor-me-diated central sensitization has been associated withthe development of hyperalgesia in conditions ofchronic pain (Haley and Wilcox, 1992; Mao et al.,1995a; Wegert et al., 1997). Similarly, it has beenargued that opioid-induced abnormal pain is depen-dent upon NMDA-mediated pain facilitation (Mao etal., 1994, 1995b; Larcher et al., 1998; Laulin et al.,1998; Celerier et al., 2000). Repeated daily injectionsof i.th. morphine to rats produced tolerance to theantinociceptive effect of morphine along with thermalhyperalgesia of the hindpaws, and both of these ef-fects were prevented by concurrent injections ofMK801 (Mao et al., 1994). In another study, thesystemic coadministration of morphine with MK801to rats prevented in a dose-dependent manner thedevelopment of tolerance to the antinociceptive effectof morphine (Trujillo and Akil, 1991). In these stud-ies, MK801 did not produce antinociception alone,nor did it increase antinociceptive action of morphinein nontolerant rats. Continuous i.th. infusion of mor-phine also reliably produced tolerance to its antinoci-ceptive effect, and the development of tolerance wasprevented by the coinfusion of the NMDA antagonistsMK801 or dextromethorphan (Manning et al., 1996).Furthermore, dextromethorphan blocked the manifes-tations of tolerance to the antinociceptive effects ofmorphine in rats (Elliott et al., 1995). Hyperalgesiathat was evoked by short-term administration of her-oin or fentanyl was blocked by the injection of theNMDA antagonists MK-801 or ketamine (Laulin etal., 1998; Celerier et al., 1999, 2000). Likewise, hy-

136 Ossipov et al.

peralgesia provoked by naloxone after heroin injec-tion was also blocked by MK-801 (Larcher et al.,1998; Laulin et al., 1999). The presence of presynap-tic NMDA receptors on central terminals of primaryafferent fibers has been demonstrated, suggesting ananatomical link between opioid and NMDA receptorsystems (Liu et al., 1994, 1997). Based on theseobservations, it may be speculated that increasedNMDA receptor activity mediates spinal sensitiza-tion, resulting in increased nociceptive input.

Dynorphin Antiserum. Manipulations that block thepronociceptive action of dynorphin also abolish ab-normal pain. A single spinal injection dynorphin hasproduced long-lasting tactile allodynia (Vanderah etal., 1996b; Laughlin et al., 1997). The i.th. injection ofantiserum to dynorphin has blocked tactile allodyniaand thermal hyperalgesia in rats receiving continuousi.th. infusion of DAMGO (Vanderah et al., 2000). Inthe same study, dynorphin antiserum unmasked theantinociceptive effect of DAMGO still present in theanimal (Vanderah et al., 2000). Moreover, the antino-ciceptive dose-response curve for spinal morphinewas restored to equal potency with that of nontolerantrats (Vanderah et al., 2000). As noted above, theenhanced evoked release of CGRP was blockedby manipulations that prevented morphine-induceddynorphin upregulation such as DLF lesion or bydynorphin antiserum (Gardell et al., 2002, 2003). In arecent study employing microdialysis, it was foundthat the introduction of NMDA, dynorphin A(1–17)or of dynorphin A(2–17) into the lumar spinal cordelicited a long-lasting release of prostaglandin E2 andof excitatory amino acids (Koetzner et al., 2004).Because these substances are associated with en-hanced sensitivity of the spinal cord to noxious inputs,this observation provides a mechanism through whichpathologically elevated levels of spinal dynorphinmay promote enhanced pain (Koetzner et al., 2004).

CGRP. Calcitonin gene-related peptide (CGRP) is aneuropeptide that has been found to be partially co-localized with substance P and glutamate in centralterminals of primary afferent neurons, and is thoughtto promote nociceptive processing in the spinal cord(Gibson et al., 1984). The ability of CGRP to antag-onize the pharmacologic actions of opioids has beendemonstrated in several studies. CGRP has been as-sociated with increased glutamate release and en-hanced NMDA receptor activity, both of which wouldoppose the antinociceptive action of opioids (Muraseet al., 1989; Kangrga et al., 1990). Exogenously ad-ministered CGRP has also elicited significant right-ward shifts in the antinociceptive dose–response

curves for morphine or [D-Pen2, D-Pen5]-enkephalin(DPDPE), and this effect was not mediated throughactions of CGRP at the opioid receptor sites (Welch etal., 1989). Long-term (up to 14 days) exposure of ratsto i.th. morphine infusion has produced antinocicep-tive tolerance along with increased spinal expressionof CGRP and a concomitant decrease in CGRP recep-tor densities (Menard et al., 1995). Importantly, therewas no corresponding increase in other spinal neu-ropeptides, including substance P, galanin, neuroten-sin, and neuropeptide Y (Menard et al., 1995).The coadministration of the CGRP antagonist,CGRP(8–37), prevented the development of antinoci-ceptive tolerance to morphine (Menard et al., 1996).Additionally, elevations in endogenous spinal CGRPlevels were prevented by this treatment as well (Me-nard et al., 1996). It was concluded that a criticalinteraction may develop between opioidergic systemsand CGRP to promote the development of tolerance toopioids (Menard et al., 1996). In a more recent study,rats were rendered tolerant to morphine by repeatedi.th. injections of morphine (Powell et al., 2000). Thecoadministration of CGRP(8–37) or of the nonpeptidicCGRP antagonist BIBN4096BS prevented the devel-opment of antinociceptive tolerance (Powell et al.,2000). More importantly, however, CGRP(8–37) re-stored the antinociceptive efficacy and potency of i.th.morphine in rats where antinociceptive tolerance wasalready established (Powell et al., 2000). The possi-bility that the apparent reversal of tolerance was dueto potentiation of morphine by CGRP(8–37) orBIBN4096BS was discounted by the fact that thesedoses alone were not antinociceptive, and that innaive rats, CGRP(8–37) slightly reduced the antinoci-ceptive action of morphine (Powell et al., 2000). It hasbeen shown that CGRP increases the release of glu-tamate and of substance P, both of which exert anexcitatory influence on postsynaptic second orderneurons of the dorsal horns of the spinal cord (Oku etal., 1987; Kangrga et al., 1990). Thus, it has beensuggested that increased CGRP release may promotenociceptive input by amplification of excitatory aminoacid activity in the spinal cord (Oku et al., 1987;Powell et al., 2000). Furthermore, CGRP given spi-nally has enhanced mechanically evoked nociceptioninduced by i.th. substance P (Kangrga et al., 1990). Asstated above, development of tolerance to morphinewas also associated with increased capsaicin-evokedrelease of CGRP from spinal cord tissue in vitro.Furthermore, spinal cord tissue taken from animalswith bilateral lesions of the DLF, which prevented thedevelopment of abnormal pain and tolerance, do notpresent with an increase in capsaicin-evoked CGRPrelease (Gardell et al., 2003). The block of descending

Actions of Opioids 137

facilitation through the DLF is believed to have pre-vented spinal plasticity leading to the manifestation ofabnormal pain along with the attendant modulation ofprimary afferent activity. It is likely that increasednociceptive input, which may be promoted in part byelevated CGRP levels, would lead to increased inputto supraspinal sites. Ultimately, activation of tonicdescending facilitation, which along with elevatedspinal dynorphin levels, would further promote en-hanced pain and tolerance to the antinociceptive ac-tion of morphine. Therefore, the development of aCGRP antagonist given along with an opioid, or of amixed CGRP antagonist/opioid agonist, is expected toprevent the development of opioid-induced abnormalpain and tolerance, and would represent a significantadvance in analgesic therapeutics.

Substance P. Because one of the primary mecha-nisms through which opioids may exert their antino-ciceptive effects is through the inhibition of release ofsubstance P from primary afferent terminals, it standsto reason that prolonged exposure to opioids may altersubstance P functionality in some fashion (Jhaman-das, 1984; Gouarderes et al., 1993). However, long-term spinal infusion with either morphine or naloxonedid not produce any changes in NK1 receptor densi-ties in the dorsal horns of the spinal cord, suggestingrather that alterations in substance P availability maybe expected instead (Gouarderes et al., 1993). Cul-tured DRG neurons that are exposed to morphine for6 days show enhanced release of substance P as wellas increased numbers of neurons immunoreactive forsubstance P (Ma et al., 2000). Likewise, long-termexposure of cultured DRG neurons to agonists actingat the mu, delta, and kappa opioid receptors produceddose-dependent increases in substance P immunore-activity (Belanger et al., 2002). Moreover, long-termspinal administration of morphine produced increasedsubstance P immunoreactivity in the spinal dorsalhorns (Powell et al., 2000). Most recently, it wasshown that coadministration of an NK1 antagonistSR140333 with spinal morphine prevented the devel-opment of antinociceptive tolerance (Powell et al.,2003). Furthermore, SR140333 reversed tolerance tospinal morphine that was firmly established (Powell etal., 2003).

COX Inhibitors. The coadministration of either ke-torolac or ibuprofen with spinal morphine over aperiod of 7 days prevented the development of an-tinociceptive tolerance to spinal morphine (Powell etal., 1999). Moreover, spinal ketorolac reversed estab-lished tolerance to morphine (Powell et al., 1999).Importantly, the doses of the COX inhibitors em-

ployed did not produce a potentiation of the antino-ciceptive effect of morphine in nontolerant animals(Powell et al., 1999). More recently, the coadminis-tration of the COX-2 inhibitor nimesulide both pre-vented and reversed antinociceptive tolerance to spi-nal morphine given over a period of 7 days (Powell etal., 2003). In addition, nimesulide prevented the up-regulation of spinal CGRP induced by persistent ex-posure to morphine (Powell et al., 2003). A complexinteraction among the NMDA receptors, nitric oxide,and PGE2 has been proposed to promote morphinetolerance (Wong et al., 2000). Nitric oxide has beenshown to directly enhance the enzymatic activity ofCOX-2, resulting in increased synthesis of PGE2(Salvemini et al., 1993; Wong et al., 2000). In turn,PGE2 may promote nitric oxide release through EP1-mediated activation of NMDA receptors, in effectpromoting a self-perpetuating sensitized state (Sakaiet al., 1998; Wong et al., 2000). These studies furthersuggest that prolonged exposure to opioids producean enhanced nociception through increased spinalsensitization.

CCK Antagonists. Cholecystokinin (CCK), de-scribed as an endogenous “antiopioid” (Stanfa et al.,1994) has been well established to have an importantrole in promoting tolerance to opioids. Substantialoverlap is seen between the distributions of CCK andof CCK receptors with those of endogenous opioidpeptides and the opioid receptors within the centralnervous system (Stengaard-Pedersen and Larsson,1981; Ghilardi et al., 1992; Verge et al., 1993). Fur-thermore, many of the sites associated with CCK arealso involved in the modulation of nociception byopiates, further strengthening the possibility of a mod-ulatory interaction between opioid-induced antinoci-ception and CCK activity. In addition, CCK coexistswith substance P in neurons of the dorsal root ganglia,where it is well poised to promote nociceptive inputsinto the spinal cord (Baber et al., 1989). The ability ofCCK to act as an endogenous antiopioid has been welldocumented (Faris et al., 1983; Watkins et al., 1985a,1985b; Kellstein et al., 1991). Exogenously adminis-tered CCK attenuates and CCK antagonists potentiate,morphine-induced antinociception in vivo and electro-physiologically (Faris et al., 1983; Stanfa et al., 1992,1994).

CCK may act as a modulator of tonic spinal an-tinociceptive activity, because morphine increases therelease of CCK in the spinal cord of the rat, whichserves to attenuate its own activity (Zhou et al., 1992,1993; Noble et al., 1993; Stanfa et al., 1994). Forexample, a single s.c. injection of morphine produceda doubling of CCK mRNA content in the hypothala-

138 Ossipov et al.

mus and a threefold increase in the spinal cord (Dingand Bayer, 1993). Moreover, prolonged exposure tomorphine has caused the upregulation of CCK in thebrain and the spinal cord, and produced a tripling ofproCCK mRNA in the hypothalamus and spinal cord,and a 97% increase in whole brain proCCK mRNA(Zhou et al., 1992; Ding and Bayer, 1993; Pu et al.,1994). Repeated injections of morphine was associ-ated with a 2.6-fold increase immunoreactive CCK inthe hypothalamus, a 2.1-fold increase in the spinalcord, and a 1.6-fold increase in the brainstem (Dingand Bayer, 1993). These results demonstrated a re-gion-specific increase in CCK content elicited bychronic morphine administration. Repeated morphineadministration also provokes an increase in CCK-likecontent from spinal cord perfusate of morphine-toler-ant rats (Zhou et al., 1993; Hoffmann and Wiesenfeld-Hallin, 1994).

Consistent with such observations, behavioral andelectrophysiological studies had shown that the CCKantagonists proglumide and lorglumide elicited anenhancement of morphine-induced antinociceptionwhile producing no antinociceptive activity whengiven alone (Watkins et al., 1985a, 1985b; Suh andTseng, 1990). Furthermore, the CCKB selective an-tagonists L365,260 and CI-988 (PD134308) enhancedthe antinociceptive effects of systemic or intrathecalmorphine (Dourish et al., 1990; Hughes et al., 1991;Ossipov et al., 1994; Vanderah et al., 1996a). Thebehavioral signs of antinociceptive tolerance to mor-phine have been reversed by CCK antiserum or pre-vented by CI-988 (Ding et al., 1986; Hoffmann andWiesenfeld-Hallin, 1994). Moreover, repeated coad-ministration of CI 988 with morphine prevented thedevelopment of antinociceptive tolerance to morphine(Xu et al., 1992). These data suggested that a truereversal of tolerance occurred, rather than a potentia-tion of the effect of morphine, because the same doseof CI988 did not potentiate morphine in drug naiverats. Increased CCK activity within the RVM mayalso drive descending facilitation of nociception (seeabove) because the direct application of antagonists tothe CCKB receptor into the RVM blocked opioid-induced abnormal pain and antinociceptive tolerance(Porreca and Vanderah, unpublished observations).Furthermore, evidence exists to show that opioid ex-posure promotes the release of CCK in the RVM(unpublished observations). These studies suggestthat persistent opioid exposure may induce neuroplas-tic changes that include increased CCK availability inthe RVM, resulting in enhanced activation of a tonicspinopetal pain facilitation arising from the RVM.This enhanced pronociceptive system elicits an up-regulation of spinal dynorphin, which serves to pro-

mote nociceptive inputs at the spinal level, and per-petuate enhanced pain through a positive feedbackcycle.

Synthesis

The opioid analgesics have been employed through-out our history for the treatment and control of pain.The opioids, exemplified by the prototype morphine,represent the most efficacious means of controllingpain at the present time. Paradoxically, the very samesubstances that are so efficacious against pain mayactually cause an abnormal pain phenomenon them-selves by eliciting the activation of endogenouspronociceptive systems. Persistent exposure to theopioids cause an increase in the presence of the prono-ciceptive neurotransmitter CCK, which acts as anendogenous modulator of antinocieptive activity. Onemeans through which this is acheived is through theactivation of a tonic descending facilitation from theRVM. Such descending facilitation may represent themissing link relating to the puzzling observation thatthe many substances which block opioid “tolerance”are substances which block excitation. Unlike thesituation with states of pain arising from inflammationor nerve injury, however, the source of such excitationfollowing prolonged exposure to opioids was un-known. It is now clear that descending pain facilita-tion arising from the RVM may represent this excita-tory input to the spinal cord, and may lead to theneuroplastic changes that could reflect a state analo-gous to the “central sensitization” characterized ininjury states. The descending pain modulatory system,spinal plasticity, and modulation of primary afferentactivity represent a point of intersection of differentparts of the nervous system and reflect mechanisms bywhich multiple mediators contribute to increased ex-citation and hyperalgesia. Critically, enhanced opioid-induced pain may require increased opioid doses toelicit antinociception or analgesia, manifesting as“tolerance.” It might be argued that increased opioiddoses would elicit a further sensitization, and thiscould explain why in long-term opioid use, doses areconstantly adjusted upward. It is also possible thatthere is a ceiling to the extent that facilitation andsensitization may occur, which would correspond toan eventual stabilization of opioid dose. Certainly,further explorations into the mechanisms driving an-tinociceptive tolerance to opioids are warrented. Animportant consequence of opioid-induced nociceptiveprocesses is that treatment of severe pain states withopioids may result in unintentional harm to patients.Understanding the neurobiology of prolonged expo-sure to opioids may allow the design of approaches to

Actions of Opioids 139

limit these adapations and may change the way inwhich opioids are used clinically.

REFERENCES

Advokat C, Burton P. 1987. Antinociceptive effect of sys-temic and intrathecal morphine in spinally transected rats.Eur J Pharmacol 139:335–343.

Aley KO, Levine JD. 1997. Different mechanisms mediatedevelopment and expression of tolerance and dependencefor peripheral mu-opioid antinociception in rat. J Neuro-sci 17:8018–8023.

Ali NM. 1986. Hyperalgesic response in a patient receivinghigh concentrations of spinal morphine. Anesthesiology65:449.

Arcaya JL, Cano G, Gomez G, Maixner W, Suarez-Roca H.1999. Dynorphin A increases substance P release fromtrigeminal primary afferent C-fibers. Eur J Pharmacol366:27–34.

Arner S, Rawal N, Gustafsson LL. 1988. Clinical experi-ence of long-term treatment with epidural and intrathecalopioids—a nationwide survey. Acta Anaesthesiol Scand32:253–259.

Arvidsson U, Dado RJ, Riedl M, Lee JH, Law PY, Loh HH,Elde R, Wessendorf MW. 1995. delta-opioid receptorimmunoreactivity: distribution in brainstem and spinalcord, and relationship to biogenic amines and enkephalin.J Neurosci 15:1215–1235.

Baber NS, Dourish CT, Hill DR. 1989. The role of CCKcaerulein, and CCK antagonists in nociception. Pain 39:307–328.

Baranauskas G, Nistri A. 1998. Sensitization of pain path-ways in the spinal cord: cellular mechanisms. Prog Neu-robiol 54:349–365.

Basbaum AI, Clanton CH, Fields HL. 1978. Three bulbospi-nal pathways from the rostral medulla of the cat: anautoradiographic study of pain modulating systems.J Comp Neurol 178:209–224.

Basbaum AI, Fields HL. 1978. Endogenous pain controlmechanisms: review and hypothesis. Ann Neurol 4:451–462.

Basbaum AI, Fields HL. 1984. Endogenous pain controlsystems: brainstem spinal pathways and endorphin cir-cuitry. Annu Rev Neurosci 7:309–338.

Basbaum AI, Marley NJ, O’Keefe J, Clanton CH. 1977.Reversal of morphine and stimulus-produced analgesiaby subtotal spinal cord lesions. Pain 3:43–56.

Bederson JB, Fields HL, Barbaro NM. 1990. Hyperalgesiaduring naloxone-precipitated withdrawal from morphineis associated with increased on-cell activity in the rostralventromedial medulla. Somatosens Mot Res 7:185–203.

Belanger S, Ma W, Chabot JG, Quirion R. 2002. Expressionof calcitonin gene-related peptide, substance P and pro-tein kinase C in cultured dorsal root ganglion neuronsfollowing chronic exposure to mu, delta and kappa opi-ates. Neuroscience 115:441–453.

Bennett GJ, Mayer DJ. 1979. Inhibition of spinal cord

interneurons by narcotic microinjection and focal electri-cal stimulation in the periaqueductal central gray matter.Brain Res 172:243–257.

Besse D, Lombard MC, Besson JM. 1991. Autoradio-graphic distribution of mu, delta and kappa opioid bind-ing sites in the superficial dorsal horn, over the rostro-caudal axis of the rat spinal cord. Brain Res 548:287–291.

Besse D, Lombard MC, Zajac JM, Roques BP, Besson JM.1990a. Pre- and postsynaptic location of mu, delta andkappa opioid receptors in the superficial layers of thedorsal horn of the rat spinal cord. Prog Clin Biol Res328:183–186.

Besse D, Lombard MC, Zajac JM, Roques BP, Besson JM.1990b. The use of (3H) DAGO to label projections of thinprimary afferent fibres at the superficial dorsal horn levelof the rat spinal cord. Prog Clin Biol Res 328:179–182.

Bian D, Ossipov MH, Ibrahim M, Raffa RB, Tallarida RJ,Malan TP Jr, Lai J, Porreca F. 1999. Loss of antiallodynicand antinociceptive spinal/supraspinal morphine synergyin nerve-injured rats: restoration by MK-801 or dynor-phin antiserum. Brain Res 831:55–63.

Burgess SE, Gardell LR, Ossipov MH, Malan TP Jr, Van-derah TW, Lai J, Porreca F. 2002. Time-dependent de-scending facilitation from the rostral ventromedial me-dulla maintains, but does not initiate, neuropathic pain.J Neurosci 22:5129–5136.

Celerier E, Laulin J, Larcher A, Le Moal M, Simonnet G.1999. Evidence for opiate-activated NMDA processesmasking opiate analgesia in rats. Brain Res 847:18–25.

Celerier E, Rivat C, Jun Y, Laulin JP, Larcher A, Reynier P,Simonnet G. 2000. Long-lasting hyperalgesia induced byfentanyl in rats: preventive effect of ketamine. Anesthe-siology 92:465–472.

Cherney NI, Portenoy RK. 1999. Practical issues in themanagement of cancer pain. In: Wall PD, Melzack R,Wall PD, Melzack RS, ediotrs. Textbook of pain.Edinburgh: Churchill Livingstone, p 1479–1522.

Childers SR. 1991. Opioid receptor-coupled second mes-senger systems. Life Sci 48:1991–2003.

Clark FM, Proudfit HK. 1993. The projections of noradren-ergic neurons in the A5 catecholamine cell group to thespinal cord in the rat: anatomical evidence that A5 neu-rons modulate nociception. Brain Res 616:200–210.

Claude P, Gracia N, Wagner L, Hargreaves KM. 1999.Effect of dynorphin on ICGRP release from capsaicin-sensitive fibers. Abstracts of the 9th World Congress onPain, vol. 9, p 262.

Collin E, Cesselin F. 1991. Neurobiological mechanisms ofopioid tolerance and dependence. Clin Neuropharmacol14:465–488.

Colpaert FC. 1978. Long-term suppression of pain by nar-cotic drugs in the absence of tolerance development. ArchInt Pharmacodyn Ther 236:293–295.

Colpaert FC. 1995. Drug discrimination: no evidence fortolerance to opiates. Pharmacol Rev 47:605–629.

Colpaert FC. 1996. System theory of pain and of opiateanalgesia: no tolerance to opiates. Pharmacol Rev 48:355–402.

140 Ossipov et al.

Colpaert FC. 2002. Mechanisms of opioid-induced pain andantinociceptive tolerance: signal transduction. Pain 95:287–288.

Colpaert FC, Kuyps JJ, Niemegeers CJ, Janssen PA. 1976.Discriminative stimulus properties of fentanyl and mor-phine: tolerance and dependence. Pharmacol BiochemBehav 5:401–408.

Conrad LC, Pfaff DW. 1976a. Efferents from medial basalforebrain and hypothalamus in the rat. I. An autoradio-graphic study of the medial preoptic area. J Comp Neurol169:185–219.

Conrad LC, Pfaff DW. 1976b. Efferents from medial basalforebrain and hypothalamus in the rat. II. An autoradio-graphic study of the anterior hypothalamus. J Comp Neu-rol 169:221–261.

Cox BM. 1990. Drug tolerance and physical dependence.In: Pratt WB, Taylor P, Pratt WB, Taylor P, editors.Principles of drug action: the basis of pharmacology.New York: Churchill Livingstone, p 639–690.

Dado RJ, Law PY, Loh HH, Elde R. 1993. Immunofluo-rescent identification of a delta (delta)-opioid receptor onprimary afferent nerve terminals. Neuroreport 5:341–344.

Dauge V, Petit F, Rossignol P, Roques BP. 1987. Use of muand delta opioid peptides of various selectivity givesfurther evidence of specific involvement of mu opioidreceptors in supraspinal analgesia (tail-flick test). EurJ Pharmacol 141:171–178.

De Conno F, Caraceni A, Martini C, Spoldi E, Salvetti M,Ventafridda V. 1991. Hyperalgesia and myoclonus withintrathecal infusion of high-dose morphine. Pain 47:337–339.

Delfs JM, Kong H, Mestek A, Chen Y, Yu L, Reisine T,Chesselet MF. 1994. Expression of mu opioid receptormRNA in rat brain: an in situ hybridization study at thesingle cell level. J Comp Neurol 345:46–68.

DePaoli AM, Hurley KM, Yasada K, Reisine T, Bell G.1994. Distribution of kappa opioid receptor mRNA inadult mouse brain: an in situ hybridization histochemistrystudy. Mol Cell Neurosci 5:327–335.

Devulder J. 1997. Hyperalgesia induced by high-dose intra-thecal sufentanil in neuropathic pain. J Neurosurg Anes-thesiol 9:146–148.

Dickenson AH, Fardin V, Le Bars D, Besson JM. 1979.Antinociceptive action following microinjection of me-thionine-enkephalin in the nucleus raphe magnus of therat. Neurosci Lett 15:265–270.

Ding XZ, Bayer BM. 1993. Increases of CCK mRNA andpeptide in different brain areas following acute andchronic administration of morphine. Brain Res 625:139–144.

Ding XZ, Fan SG, Zhou JP, Han JS. 1986. Reversal oftolerance to morphine but no potentiation of morphine-induced analgesia by antiserum against cholecystokininoctapeptide. Neuropharmacology 25:1155–1160.

Dourish CT, Hawley D, Iversen SD. 1988. Enhancement ofmorphine analgesia and prevention of morphine tolerancein the rat by the cholecystokinin antagonist L-364,718.Eur J Pharmacol 147:469–472.

Dourish CT, O’Neill MF, Coughlan J, Kitchener SJ, Haw-ley D, Iversen SD. 1990. The selective CCK-B receptorantagonist L-365,260 enhances morphine analgesia andprevents morphine tolerance in the rat. Eur J Pharmacol176:35–44.

Draisci G, Kajander KC, Dubner R, Bennett GJ, IadarolaMJ. 1991. Up-regulation of opioid gene expression inspinal cord evoked by experimental nerve injuries andinflammation. Brain Res 560:186–192.

Dubner R, Ruda MA. 1992. Activity-dependent neuronalplasticity following tissue injury and inflammation.Trends Neurosci 15:96–103.

Duggan AW, Hall JG, Headley PM. 1976. Morphine, en-kephalin and the substantia gelatinosa. Nature 264:456–458.

Dworkin RH, Backonja M, Rowbotham MC, Allen RR,Argoff CR, Bennett GJ, Bushnell MC, Farrar JT, GalerBS, Haythornthwaite JA, et al. 2003. Advances in neu-ropathic pain: diagnosis, mechanisms, and treatment rec-ommendations. Arch Neurol 60:1524–1534.

Eisenach JC. 2000. Preemptive hyperalgesia, not analgesia?Anesthesiology 92:308–309.

Elliott KJ, Brodsky M, Hyanansky A, Foley KM, InturrisiCE. 1995. Dextromethorphan shows efficacy in experi-mental pain (nociception) and opioid tolerance. Neurol-ogy 45:S66–S68.

Erspamer V, Melchiorri P, Falconieri-Erspamer G, Negri L,Corsi R, Severini C, Barra D, Simmaco M, Kreil G. 1989.Deltorphins: a family of naturally occurring peptides withhigh affinity and selectivity for delta opioid binding sites.Proc Natl Acad Sci USA 86:5188–5192.

Faden AI. 1992. Dynorphin increases extracellular levels ofexcitatory amino acids in the brain through a non-opioidmechanism. J Neurosci 12:425–429.

Faris PL, Komisaruk BR, Watkins LR, Mayer DJ. 1983.Evidence for the neuropeptide cholecystokinin as an an-tagonist of opiate analgesia. Science 219:310–312.

Fernandes M, Kluwe S, Coper H. 1977a. The developmentof tolerance to morphine in the rat. Psychopharmacology54:197–201.

Fernandes M, Kluwe S, Coper H. 1977b. Quantitative as-sessment of tolerance to and dependence on morphine inmice. Naunyn Schmiedebergs Arch Pharmacol 297:53–60.

Fields HL. 1992. Is there a facilitating component to centralpain modulation? APS J 1:71–78.

Fields HL. 2000. Pain modulation: expectation, opioid an-algesia and virtual pain. Prog Brain Res 122:245–253.

Fields HL, Anderson SD. 1978. Evidence that raphe-spinalneurons mediate opiate and midbrain stimulation-pro-duced analgesias. Pain 5:333–349.

Fields HL, Barbaro NM, Heinricher MM. 1988. Brain stemneuronal circuitry underlying the antinociceptive actionof opiates. Prog Brain Res 77:245–257.

Fields HL, Basbaum AI. 1978. Brainstem control of spinalpain-transmission neurons. Annu Rev Physiol 40:217–248.

Fields HL, Basbaum AI. 1999. Central nervous system

Actions of Opioids 141

mechanisms of pain modulation. In: Wall PD, Melzack R,Wall PD, Melzack R, editors. Textbook of pain. Edin-burgh: Churchill Livingstone, p 309–329.

Fields HL, Bry J, Hentall I, Zorman G. 1983. The activity ofneurons in the rostral medulla of the rat during with-drawal from noxious heat. J Neurosci 3:2545–2552.

Fields HL, Heinricher MM. 1985. Anatomy and physiologyof a nociceptive modulatory system. Philos Trans R SocLond B Biol Sci 308:361–374.

Foley KM. 1993. Opioids. Neurol Clin 11:503–522.Foley KM. 1995. Misconceptions and controversies regard-

ing the use of opioids in cancer pain. Anticancer Drugs6(Suppl 3):4–13.

Gardell LR, Vanderah TW, Gardell SE, Wang R, OssipovMH, Lai J, Porreca F. 2003. Enhanced evoked excitatorytransmitter release in experimental neuropathy requiresdescending facilitation. J Neurosci 23:8370–8379.

Gardell LR, Wang R, Burgess SE, Ossipov MH, VanderahTW, Malan TP Jr, Lai J, Porreca F. 2002. Sustainedmorphine exposure induces a spinal dynorphin-dependentenhancement of excitatory transmitter release from pri-mary afferent fibers. J Neurosci 22:6747–6755.

Gebhart GF, Ossipov MH. 1986. Characterization of inhi-bition of the spinal nociceptive tail-flick reflex in the ratfrom the medullary lateral reticular nucleus. J Neurosci6:701–713.

Ghilardi JR, Allen CJ, Vigna SR, McVey DC, Mantyh PW.1992. Trigeminal and dorsal root ganglion neurons ex-press CCK receptor binding sites in the rat, rabbit, andmonkey: possible site of opiate–CCK analgesic interac-tions. J Neurosci 12:4854–4866.

Gibson SJ, Polak JM, Bloom SR, Sabate IM, Mulderry PM,Ghatei MA, McGregor GP, Morrison JF, Kelly JS, EvansRM, et al. 1984. Calcitonin gene-related peptide immu-noreactivity in the spinal cord of man and of eight otherspecies. J Neurosci 4:3101–3111.

Gouarderes C, Jhamandas K, Cridland R, Cros J, Quirion R,Zajac JM. 1993. Opioid and substance P receptor adap-tations in the rat spinal cord following sub-chronic intra-thecal treatment with morphine and naloxone. Neuro-science 54:799–807.

Gutstein HB. 1996. The effects of pain on opioid tolerance:how do we resolve the controversy? Pharmacol Rev48:403–407.

Gutstein HB, Trujillo KA, Akil H. 1995. Does chronicnociceptive stimulation alter the development of mor-phine tolerance? Brain Res 680:173–179.

Haley JE, Wilcox GL. 1992. Involvement of excitatoryamino acids and peptides in the spinal mechanisms un-derlying hyperalgesia. In: Willis WD, editor. Hyperalge-sia and allodynia. New York: Raven Press, p 281–293.

Heinricher MM, McGaraughty S. 1996. CCK modulates theantinociceptive actions of opioids by an action within therostral ventromedial medulla: a combined electrophysio-logical and behavioral study. In: Abstracts of the 8thWorld Congress on Pain, Vancouver. Seattle: IASP Press,p 472.

Heinricher MM, Pertovaara A, Ossipov MH. 2003. De-

scending modulation after injury. In: Dostrovsky DO,Carr DB, Koltzenburg M, Dostrovsky DO, Carr DB,Koltzenburg M, editors. Proceedings of the 10th WorldCongress on Pain. Seattle: IASP Press, p 251–260.

Heinricher MM, Roychowdhury SM. 1997. Reflex-relatedactivation of putative pain facilitating neurons in rostralventromedial medulla requires excitatory amino acidtransmission. Neuroscience 78:1159–1165.

Hoffmann O, Wiesenfeld-Hallin Z. 1994. The CCK-B re-ceptor antagonist Cl 988 reverses tolerance to morphinein rats. Neuroreport 5:2565–2568.

Hughes J, Hunter JC, Woodruff GN. 1991. Neurochemicalactions of CCK underlying the therapeutic potential ofCCK-B antagonists. Neuropeptides 19(Suppl):85–89.

Jacquet YF, Lajtha A. 1973. Morphine action at centralnervous system sites in rat: analgesia or hyperalgesiadepending on site and dose. Science 182:490–492.

Jensen TS, Yaksh TL. 1986a. Comparison of antinocicep-tive action of morphine in the periaqueductal gray, medialand paramedial medulla in rat. Brain Res 363:99–113.

Jensen TS, Yaksh TL. 1986b. Comparison of the antinoci-ceptive action of mu and delta opioid receptor ligands inthe periaqueductal gray matter, medial and paramedialventral medulla in the rat as studied by the microinjectiontechnique. Brain Res 372:301–312.

Jhamandas KH. 1984. Opioid-neurotransmitter interactions:significance in analgesia, tolerance and dependence. ProgNeuropsychopharmacol Biol Psychiatry 8:565–570.

Ji RR, Zhang Q, Law PY, Low HH, Elde R, Hokfelt T.1995. Expression of mu-, delta-, and kappa-opioid recep-tor-like immunoreactivities in rat dorsal root ganglia aftercarrageenan-induced inflammation. J Neurosci 15:8156–8166.

Jiang Q, Mosberg HI, Porreca F. 1990. Antinociceptiveeffects of [D-Ala2]deltorphin II, a highly selective deltaagonist in vivo. Life Sci 47:PL43–PL47.

Jones SL, Gebhart GF. 1986a. Characterization of coeru-leospinal inhibition of the nociceptive tail-flick reflex inthe rat: mediation by spinal alpha 2-adrenoceptors. BrainRes 364:315–330.

Jones SL, Gebhart GF. 1986b. Quantitative characterizationof ceruleospinal inhibition of nociceptive transmission inthe rat. J Neurophysiol 56:1397–1410.

Jurna I, Grossman W. 1976. The effect of morphine on theactivity evoked in ventrolateral tract axons of the catspinal cord. Exp Brain Res 24:473–484.

Kajander KC, Sahara Y, Iadarola MJ, Bennett GJ. 1990.Dynorphin increases in the dorsal spinal cord in rats witha painful peripheral neuropathy. Peptides 11:719–728.

Kalyuzhny AE, Arvidsson U, Wu W, Wessendorf MW.1996. mu-Opioid and delta-opioid receptors are ex-pressed in brainstem antinociceptive circuits: studies us-ing immunocytochemistry and retrograde tract-tracing.J Neurosci 16:6490–6503.

Kangrga I, Larew JS, Randic M. 1990. The effects ofsubstance P and calcitonin gene-related peptide on theefflux of endogenous glutamate and aspartate from the ratspinal dorsal horn in vitro. Neurosci Lett 108:155–160.

142 Ossipov et al.

Kaplan H, Fields HL. 1991. Hyperalgesia during acuteopioid abstinence: evidence for a nociceptive facilitatingfunction of the rostral ventromedial medulla. J Neurosci11:1433–1439.

Kayser V, Guilbaud G. 1985. Can tolerance to morphine beinduced in arthritic rats? Brain Res 334:335–338.

Kayser V, Neil A, Guilbaud G. 1986. Repeated low doses ofmorphine induce a rapid tolerance in arthritic rats but apotentiation of opiate analgesia in normal animals. BrainRes 383:392–396.

Kieffer BL, Evans CJ. 2002. Opioid tolerance—in search ofthe holy grail. Cell 108:587–590.

Kellstein DE, Price DD, Mayer DJ. 1991. Cholecystokininand its antagonist lorglumide respectively attenuate andfacilitate morphine-induced inhibition of C-fiber evokeddischarges of dorsal horn nociceptive neurons. Brain Res540:302–306.

Kest B, McLemore G, Kao B, Inturrisi CE. 1997. Thecompetitive alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate receptor antagonist LY293558 attenuatesand reverses analgesic tolerance to morphine but not todelta or kappa opioids. J Pharmacol Exp Ther 283:1249–1255.

Kiefel JM, Rossi GC, Bodnar RJ. 1993. Medullary mu anddelta opioid receptors modulate mesencephalic morphineanalgesia in rats. Brain Res 624:151–161.

Kim DH, Fields HL, Barbaro NM. 1990. Morphine analge-sia and acute physical dependence: rapid onset of twoopposing, dose-related processes. Brain Res 516:37–40.

Koetzner L, Hua XY, Lai J, Porreca F, Yaksh T. 2004.Nonopioid actions of intrathecal dynorphin evoke spinalexcitatory amino acid and prostaglandin E2 release me-diated by cyclooxygenase-1 and -2. J Neurosci 24:1451–1458.

Kovelowski CJ, Ossipov MH, Sun H, Lai J, Malan TP,Porreca F. 2000. Supraspinal cholecystokinin may drivetonic descending facilitation mechanisms to maintainneuropathic pain in the rat. Pain 87:265–273.

Lai J, Riedl M, Stone LS, Arvidsson U, Bilsky EJ, WilcoxGL, Elde R, Porreca F. 1996. Immunofluorescence anal-ysis of antisense oligodeoxynucleotide-mediated “knock-down” of the mouse delta opioid receptor in vitro and invivo. Neurosci Lett 213:205–208.

Larcher A, Laulin JP, Celerier E, Le Moal M, Simonnet G.1998. Acute tolerance associated with a single opiateadministration: involvement of N-methyl-D-aspartate-de-pendent pain facilitatory systems. Neuroscience 84:583–589.

Laughlin TM, Vanderah TW, Lashbrook J, Nichols ML,Ossipov M, Porreca F, Wilcox GL. 1997. Spinally ad-ministered dynorphin A produces long-lasting allodynia:involvement of NMDA but not opioid receptors. Pain72:253–260.

Laulin JP, Celerier E, Larcher A, Le Moal M, Simonnet G.1999. Opiate tolerance to daily heroin administration: anapparent phenomenon associated with enhanced pain sen-sitivity. Neuroscience 89:631–636.

Laulin JP, Larcher A, Celerier E, Le Moal M, Simonnet G.

1998. Long-lasting increased pain sensitivity in rat fol-lowing exposure to heroin for the first time. Eur J Neu-rosci 10:782–785.

Le Bars D, Menetrey D, Conseiller C, Besson JM. 1975.Depressive effects of morphine upon lamina V cellsactivities in the dorsal horn of the spinal cat. Brain Res98:261–277.

Lewis VA, Gebhart GF. 1977a. Evaluation of the periaque-ductal central gray (PAG) as a morphine-specific locus ofaction and examination of morphine-induced and stimu-lation-produced analgesia at coincident PAG loci. BrainRes 124:283–303.

Lewis VA, Gebhart GF. 1977b. Morphine-induced andstimulation-produced analgesias at coincident periaque-ductal central gray loci: evaluation of analgesic congru-ence, tolerance, and cross-tolerance. Exp Neurol 57:934–955.

Liu H, Mantyh PW, Basbaum AI. 1997. NMDA-receptorregulation of substance P release from primary afferentnociceptors. Nature 386:721–724.

Liu H, Wang H, Sheng M, Jan LY, Jan YN, Basbaum AI.1994. Evidence for presynaptic N-methyl-D-aspartate au-toreceptors in the spinal cord dorsal horn. Proc Natl AcadSci USA 91:8383–8387.

Lombard MC, Besse D, Besson JM. 1995. Opioid receptorsin the superficial layers of the rat spinal cord: functionalimplications in pain processing. Prog Brain Res 104:77–92.

Lord JA, Waterfield AA, Hughes J, Kosterlitz HW. 1977.Endogenous opioid peptides: multiple agonists and recep-tors. Nature 267:495–499.

Lutfy K, Shen KZ, Woodward RM, Weber E. 1996. Inhi-bition of morphine tolerance by NMDA receptor antag-onists in the formalin test. Brain Res 731:171–181.

Ma QP, Woolf CJ. 1995. Noxious stimuli induce an N-methyl-D-aspartate receptor-dependent hypersensitivityof the flexion withdrawal reflex to touch: implications forthe treatment of mechanical allodynia. Pain 61:383–390.

Ma W, Zheng WH, Kar S, Quirion R. 2000. Morphinetreatment induced calcitonin gene-related peptide andsubstance P increases in cultured dorsal root ganglionneurons. Neuroscience 99:529–539.

Malan TP, Ossipov MH, Gardell LR, Ibrahim M, Bian D,Lai J, Porreca F. 2000. Extraterritorial neuropathic paincorrelates with multisegmental elevation of spinal dynor-phin in nerve-injured rats. Pain 86:185–194.

Manning BH, Mao J, Frenk H, Price DD, Mayer DJ. 1996.Continuous co-administration of dextromethorphan orMK-801 with morphine: attenuation of morphine depen-dence and naloxone-reversible attenuation of morphinetolerance. Pain 67:79–88.

Mansour A, Fox CA, Akil H, Watson SJ. 1995. Opioid-receptor mRNA expression in the rat CNS: anatomicaland functional implications. Trends Neurosci 18:22–29.

Mansour A, Fox CA, Burke S, Meng F, Thompson RC, AkilH, Watson SJ. 1994a. Mu, delta, and kappa opioid recep-tor mRNA expression in the rat CNS: an in situ hybrid-ization study. J Comp Neurol 350:412–438.

Actions of Opioids 143

Mansour A, Fox CA, Thompson RC, Akil H, Watson SJ.1994b. mu-Opioid receptor mRNA expression in the ratCNS: comparison to mu-receptor binding. Brain Res 643:245–265.

Mansour A, Khachaturian H, Lewis ME, Akil H, WatsonSJ. 1987. Autoradiographic differentiation of mu, delta,and kappa opioid receptors in the rat forebrain and mid-brain. J Neurosci 7:2445–2464.

Mansour A, Thompson RC, Akil H, Watson SJ. 1993. Deltaopioid receptor mRNA distribution in the brain: compar-ison to delta receptor binding and proenkephalin mRNA.J Chem Neuroanat 6:351–362.

Mao J, Price DD, Caruso FS, Mayer DJ. 1996. Oral admin-istration of dextromethorphan prevents the developmentof morphine tolerance and dependence in rats. Pain 67:361–368.

Mao J, Price DD, Hayes RL, Lu J, Mayer DJ, Frenk H.1993. Intrathecal treatment with dextrorphan or ketaminepotently reduces pain-related behaviors in a rat model ofperipheral mononeuropathy. Brain Res 605:164–168.

Mao J, Price DD, Lu J, Mayer DJ. 1998. Antinociceptivetolerance to the mu-opioid agonist DAMGO is dose-dependently reduced by MK-801 in rats. NeuroscienceLetters 250:193–196.

Mao J, Price DD, Mayer DJ. 1994. Thermal hyperalgesia inassociation with the development of morphine tolerancein rats: roles of excitatory amino acid receptors andprotein kinase C. J Neurosci 14:2301–2312.

Mao J, Price DD, Mayer DJ. 1995a. Experimental monon-europathy reduces the antinociceptive effects of mor-phine: implications for common intracellular mechanismsinvolved in morphine tolerance and neuropathic pain.Pain 61:353–364.

Mao J, Price DD, Mayer DJ. 1995b. Mechanisms of hyper-algesia and morphine tolerance: a current view of theirpossible interactions. Pain 62:259–274.

Mao J, Price DD, Mayer DJ, Hayes RL. 1992a. Pain-relatedincreases in spinal cord membrane-bound protein kinaseC following peripheral nerve injury. Brain Res 588:144–149.

Mao J, Price DD, Mayer DJ, Lu J, Hayes RL. 1992b.Intrathecal MK-801 and local nerve anesthesia synergis-tically reduce nociceptive behaviors in rats with experi-mental peripheral mononeuropathy. Brain Res 576:254–262.

Mao J, Price DD, Phillips LL, Lu J, Mayer DJ. 1995c.Increases in protein kinase C gamma immunoreactivity inthe spinal cord of rats associated with tolerance to theanalgesic effects of morphine. Brain Res 677:257–267.

Martin WR. 1983. Pharmacology of opioids. PharmacolRev 35:283–323.

Martin WR, Eades CG, Thompson JA, Huppler RE, GilbertPE. 1976. The effects of morphine- and nalorphine-likedrugs in the nondependent and morphine-dependentchronic spinal dog. J Pharmacol Exp Ther 197:517–532.

Massotte D, Kieffer BL. 1998. A molecular basis for opiateaction. Essays Biochem 33:65–77.

Mayer DJ, Mao J, Price DD. 1995a. The association of

neuropathic pain, morphine tolerance and dependence,and the translocation of protein kinase C. NIDA ResMonogr 147:269–298.

Mayer DJ, Mao J, Price DD. 1995b. The development ofmorphine tolerance and dependence is associated withtranslocation of protein kinase C. Pain 61:365–374.

McClane TK, Martin WR. 1967. Effects of morphine, na-lorphine, cyclazocine, and naloxone on the flexor reflex.Int J Neuropharmacol 6:89–98.

McGowan MK, Hammond DL. 1993. Antinociception pro-duced by microinjection of L-glutamate into the ventro-medial medulla of the rat: mediation by spinal GABAAreceptors. Brain Res 620:86–96.

McMahon SB, Wall PD. 1983. A system of rat spinal cordlamina 1 cells projecting through the contralateral dorso-lateral funiculus. J Comp Neurol 214:217–223.

McMahon SB, Wall PD. 1988. Descending excitation andinhibition of spinal cord lamina I projection neurons.J Neurophysiol 59:1204–1219.

McNally GP. 1999. Pain facilitatory circuits in the mam-malian central nervous system: their behavioral signifi-cance and role in morphine analgesic tolerance. NeurosciBiobehav Rev 23:1059–1078.

Menard DP, van Rossum D, Kar S, Jolicoeur FB, Jhaman-das K, Quirion R. 1995. Tolerance to the antinociceptiveproperties of morphine in the rat spinal cord: alteration ofcalcitonin gene-related peptide-like immunostaining andreceptor binding sites. J Pharmacol Exp Ther 273:887–894.

Menard DP, van Rossum D, Kar S, St Pierre S, Sutak M,Jhamandas K, Quirion R. 1996. A calcitonin gene-relatedpeptide receptor antagonist prevents the development oftolerance to spinal morphine analgesia. J Neurosci 16:2342–2351.

Miaskowski C, Taiwo YO, Levine JD. 1991. Contributionof supraspinal mu- and delta-opioid receptors to antino-ciception in the rat. Eur J Pharmacol 205:247–252.

Minami M, Hosoi Y, Toya T, Katao Y, Maekawa K, Kat-sumata S, Yabuuchi K, Onogi T, Satoh M. 1993a. In situhybridization study of kappa-opioid receptor mRNA inthe rat brain. Neurosci Lett 162:161–164.

Minami M, Onogi T, Toya T, Katao Y, Hosoi Y, MaekawaK, Katsumata S, Yabuuchi K, Satoh M. 1994. Molecularcloning and in situ hybridization histochemistry for ratmu-opioid receptor. Neurosci Res 18:315–322.

Minami M, Toya T, Katao Y, Maekawa K, Nakamura S,Onogi T, Kaneko S, Satoh M. 1993b. Cloning and ex-pression of a cDNA for the rat kappa-opioid receptor.FEBS Lett 329:291–295.

Morgan MM, Fields HL. 1994. Pronounced changes in theactivity of nociceptive modulatory neurons in the rostralventromedial medulla in response to prolonged thermalnoxious stimuli. J Neurophysiol 72:1161–1170.

Murase K, Ryu PD, Randic M. 1989. Excitatory and inhib-itory amino acids and peptide-induced responses inacutely isolated rat spinal dorsal horn neurons. NeurosciLett 103:56–63.

Neil A, Kayser V, Chen YL, Guilbaud G. 1990. Repeated

144 Ossipov et al.

low doses of morphine do not induce tolerance but in-crease the opioid antinociceptive effect in rats with aperipheral neuropathy. Brain Res 522:140–143.

Noble F, Derrien M, Roques BP. 1993. Modulation ofopioid antinociception by CCK at the supraspinal level:evidence of regulatory mechanisms between CCK andenkephalin systems in the control of pain. Br J Pharmacol109:1064–1070.

Oku R, Satoh M, Fujii N, Otaka A, Yajima H, Takagi H.1987. Calcitonin gene-related peptide promotes mechan-ical nociception by potentiating release of substance Pfrom the spinal dorsal horn in rats. Brain Res 403:350–354.

Oliveras JL, Hosobuchi Y, Redjemi F, Guilbaud G, BessonJM. 1977. Opiate antagonist, naloxone, strongly reducesanalgesia induced by stimulation of a raphe nucleus (cen-tralis inferior). Brain Res 120:221–229.

Ossipov MH, Hong Sun T, Malan P, Jr., Lai J, Porreca F.2000a. Mediation of spinal nerve injury induced tactileallodynia by descending facilitatory pathways in the dor-solateral funiculus in rats. Neurosci Lett 290:129–132.

Ossipov MH, Kovelowski CJ, Nichols ML, Hruby VJ,Porreca F. 1995a. Characterization of supraspinal antino-ciceptive actions of opioid delta agonists in the rat. Pain62:287–293.

Ossipov MH, Kovelowski CJ, Porreca F. 1995b. The in-crease in morphine antinociceptive potency produced bycarrageenan-induced hindpaw inflammation is blockedby naltrindole, a selective delta-opioid antagonist. Neu-rosci Lett 184:173–176.

Ossipov MH, Kovelowski CJ, Vanderah T, Porreca F. 1994.Naltrindole, an opioid delta antagonist, blocks the en-hancement of morphine-antinociception induced by aCCKB antagonist in the rat. Neurosci Lett 181:9–12.

Ossipov MH, Lai J, Malan TP, Jr., Porreca F. 2000b. Spinaland supraspinal mechanisms of neuropathic pain. Ann NY Acad Sci 909:12–24.

Ossipov MH, Lai J, Malan TP Jr, Vanderah TW, Porreca F.2001. Tonic descending facilitation as a mechanism ofneuropathic pain. In: Hansson PT, Fields HL, Hill RG,Marchettini P, Hansson PT, Fields HL, Hill RG, Mar-chettini P, editors. Neuropatic pain: pathophysiology andtreatment. Seattle: IASP Press, p 107–124.

Ossipov MH, Lai J, Malan TP, Porreca F. 1998. Recentadvances in the pharmacology of opioids. In: Sawynok J,Cowan A, Sawynok J, Cowan A, editors. Novel aspectsof pain management: opioids and beyond. New York:John Wiley & Sons.

Ossipov MH, Lai J, Vanderah TW, Porreca F. 2004. Theopioid delta receptor in pain modulation. In: Chang kj,Porreca F, Woods JH, Chang kj, Porreca F, Woods JH,editors. The delta receptor. New York: Marcel Dekker,Inc., p 297–329.

Pert CB, Snyder SH. 1973a. Opiate receptor: demonstrationin nervous tissue. Science 179:1011–1014.

Pert CB, Snyder SH. 1973b. Properties of opiate-receptorbinding in rat brain. Proc Natl Acad Sci USA 70:2243–2247.

Pertovaara A, Wei H, Hamalainen MM. 1996. Lidocaine inthe rostroventromedial medulla and the periaqueductalgray attenuates allodynia in neuropathic rats. NeurosciLett 218:127–130.

Pohl M, Ballet S, Collin E, Mauborgne A, Bourgoin S,Benoliel JJ, Hamon M, Cesselin F. 1997. Enkephalin-ergic and dynorphinergic neurons in the spinal cord anddorsal root ganglia of the polyarthritic rat—in vivo re-lease and cDNA hybridization studies. Brain Res 749:18–28.

Pol O, Puig MM. 1997. Reversal of tolerance to the anti-transit effects of morphine during acute intestinal inflam-mation in mice. Br J Pharmacol 122:1216–1222.

Porreca F, Ossipov MH, Gebhart GF. 2002. Chronic painand medullary descending facilitation. Trends Neurosci25:319–325.

Porreca F, Burgess SE, Gardell LR, Vanderah TW, MalanTP Jr, Ossipov MH, Lappi DA, Lai J. 2001. Inhibition ofneuropathic pain by selective ablation of brainstem med-ullary cells expressing the mu-opioid receptor. J Neurosci21:5281–5288.

Porreca F, Mosberg HI, Hurst R, Hruby VJ, Burks TF.1984. Roles of mu, delta and kappa opioid receptors inspinal and supraspinal mediation of gastrointestinal tran-sit effects and hot-plate analgesia in the mouse. J Phar-macol Exp Ther 230:341–348.

Powell KJ, Hosokawa A, Bell A, Sutak M, Milne B, QuirionR, Jhamandas K. 1999. Comparative effects of cyclo-oxygenase and nitric oxide synthase inhibition on thedevelopment and reversal of spinal opioid tolerance. Br JPharmacol 127:631–644.

Powell KJ, Ma W, Sutak M, Doods H, Quirion R, Jhaman-das K. 2000. Blockade and reversal of spinal morphinetolerance by peptide and non-peptide calcitonin gene-related peptide receptor antagonists. Br J Pharmacol 131:875–884.

Powell KJ, Quirion R, Jhamandas K. 2003. Inhibition ofneurokinin-1-substance P receptor and prostanoid activityprevents and reverses the development of morphine tol-erance in vivo and the morphine-induced increase inCGRP expression in cultured dorsal root ganglion neu-rons. Eur J Neurosci 18:1572–1583.

Proudfit HK, Hammond DL. 1981. Alterations in nocicep-tive threshold and morphine-induced analgesia producedby intrathecally administered amine antagonists. BrainRes 218:393–399.

Pu S, Zhuang H, Lu Z, Wu X, Han J. 1994. Cholecystokiningene expression in rat amygdaloid neurons: normal dis-tribution and effect of morphine tolerance. Brain Res MolBrain Res 21:183–189.

Quirion R. 1984. Pain, nociception and spinal opioidreceptors. Prog Neuropsychopharmacol Biol Psychia-try 8:571–579.

Quirion R, Zajac JM, Morgat JL, Roques BP. 1983. Auto-radiographic distribution of mu and delta opiate receptorsin rat brain using highly selective ligands. Life Sci33(Suppl 1):227–230.

Roerig SC, Fujimoto JM. 1988. Morphine antinociception

Actions of Opioids 145

in different strains of mice: relationship of supraspinal-spinal multiplicative interaction to tolerance. J PharmacolExp Ther 247:603–608.

Roerig SC, Fujimoto JM. 1989. Multiplicative interactionbetween intrathecally and intracerebroventricularly ad-ministered mu opioid agonists but limited interactionsbetween delta and kappa agonists for antinociception inmice. J Pharmacol Exp Ther 249:762–768.

Rossi GC, Pasternak GW, Bodnar RJ. 1993. Synergisticbrainstem interactions for morphine analgesia. Brain Res624:171–180.

Rowbotham MC. 1995. Chronic pain: from theory to prac-tical management. Neurology 45:S5–S10; discussionS35–S16.

Rowbotham MC, Twilling L, Davies PS, Reisner L, TaylorK, Mohr D. 2003. Oral opioid therapy for chronic periph-eral and central neuropathic pain. N Engl J Med 348:1223–1232.

Sabbe MB, Yaksh TL. 1990. Pharmacology of spinal opi-oids. J Pain Symptom Manage 5:191–203.

Sakai M, Minami T, Hara N, Nishihara I, Kitade H, Ka-miyama Y, Okuda K, Takahashi H, Mori H, Ito S. 1998.Stimulation of nitric oxide release from rat spinal cord byprostaglandin E2. Br J Pharmacol 123:890–894.

Salvemini D, Misko TP, Masferrer JL, Seibert K, CurrieMG, Needleman P. 1993. Nitric oxide activates cycloox-ygenase enzymes. Proc Natl Acad Sci USA 90:7240–7244.

Satoh M, Minami M. 1995. Molecular pharmacology of theopioid receptors. Pharmacol Ther 68:343–364.

Schmidt P, Schroder H, Maderspach K, Staak M. 1994.Immunohistochemical localization of kappa opioid recep-tors in the human frontal cortex. Brain Res 654:223–233.

Schulz R, Wuster M, Herz A. 1981. Pharmacological char-acterization of the epsilon-opiate receptor. J PharmacolExp Ther 216:604–606.

Simon EJ. 1973. In search of the opiate receptor. Am J MedSci 266:160–168.

Siuciak JA, Advokat C. 1989. The synergistic effect ofconcurrent spinal and supraspinal opiate agonisms is re-duced by both nociceptive and morphine pretreatment.Pharmacol Biochem Behav 34:265–273.

Skilling SR, Sun X, Kurtz HJ, Larson AA. 1992. Selectivepotentiation of NMDA-induced activity and release ofexcitatory amino acids by dynorphin: possible roles inparalysis and neurotoxicity. Brain Res 575:272–278.

Stanfa L, Dickenson A, Xu XJ, Wiesenfeld-Hallin Z. 1994.Cholecystokinin and morphine analgesia: variations on atheme. Trends Pharmacol Sci 15:65–66.

Stanfa LC, Sullivan AF, Dickenson AH. 1992. Alterationsin neuronal excitability and the potency of spinal mu,delta and kappa opioids after carrageenan-induced in-flammation. Pain 50:345–354.

Stein C, Pfluger M, Yassouridis A, Hoelzl J, Lehrberger K,Welte C, Hassan AH. 1996. No tolerance to peripheralmorphine analgesia in presence of opioid expression ininflamed synovia. J Clin Invest 98:793–799.

Stengaard-Pedersen K, Larsson LI. 1981. Comparative im-

munocytochemical localization of putative opioid ligandsin the central nervous system. Histochemistry 73:89–114.

Stillman MJ, Moulin DE, Foley KM. 1987. Paradoxicalpain following high-dose spinal morphine. Pain 4.

Suh HH, Tseng LF. 1990. Differential effects of sulfatedcholecystokinin octapeptide and proglumide injected in-trathecally on antinociception induced by beta-endorphinand morphine administered intracerebroventricularly inmice. Eur J Pharmacol 179:329–338.

Svingos AL, Moriwaki A, Wang JB, Uhl GR, Pickel VM.1996. Ultrastructural immunocytochemical localizationof mu-opioid receptors in rat nucleus accumbens: extra-synaptic plasmalemmal distribution and association withLeu5-enkephalin. J Neurosci 16:4162–4173.

Taddese A, Nah SY, McCleskey EW. 1995. Selective opi-oid inhibition of small nociceptive neurons. Science 270:1366–1369.

Terenius L. 1973. Stereospecific interaction between nar-cotic analgesics and a synaptic plasm a membrane frac-tion of rat cerebral cortex. Acta Pharmacol Toxicol(Copenh) 32:317–320.

Tiseo PJ, Cheng J, Pasternak GW, Inturrisi CE. 1994. Mod-ulation of morphine tolerance by the competitive N-methyl-D-aspartate receptor antagonist LY274614: as-sessment of opioid receptor changes. J Pharmacol ExpTher 268:195–201.

Tiseo PJ, Inturrisi CE. 1993. Attenuation and reversal ofmorphine tolerance by the competitive N-methyl-D-as-partate receptor antagonist, LY274614. J Pharmacol ExpTher 264:1090–1096.

Trujillo KA, Akil H. 1991. Inhibition of morphine toleranceand dependence by the NMDA receptor antagonist MK-801. Science 251:85–87.

Trujillo KA, Akil H. 1994. Inhibition of opiate tolerance bynon-competitive N-methyl-D-aspartate receptor antago-nists. Brain Res 633:178–188.

Urban MO, Gebhart GF. 1999. Supraspinal contributions tohyperalgesia. Proc Natl Acad Sci USA 96:7687–7692.

Van Bockstaele EJ, Colago EE, Cheng P, Moriwaki A, UhlGR, Pickel VM. 1996. Ultrastructural evidence for prom-inent distribution of the mu-opioid receptor at extrasyn-aptic sites on noradrenergic dendrites in the rat nucleuslocus coeruleus. J Neurosci 16:5037–5048.

Vanderah TW, Bernstein RN, Yamamura HI, Hruby VJ,Porreca F. 1996a. Enhancement of morphine antinocicep-tion by a CCKB antagonist in mice is mediated via opioiddelta receptors. J Pharmacol Exp Ther 278:212–219.

Vanderah TW, Gardell LR, Burgess SE, Ibrahim M, DogrulA, Zhong CM, Zhang ET, Malan TP Jr, Ossipov MH, LaiJ, et al. 2000. Dynorphin promotes abnormal pain andspinal opioid antinociceptive tolerance. J Neurosci 20:7074–7079.

Vanderah TW, Laughlin T, Lashbrook JM, Nichols ML,Wilcox GL, Ossipov MH, Malan TP Jr, Porreca F. 1996b.Single intrathecal injections of dynorphin A or des-Tyr-dynorphins produce long-lasting allodynia in rats: block-ade by MK-801 but not naloxone. Pain 68:275–281.

Vanderah TW, Ossipov MH, Lai J, Malan TP Jr, Porreca F.

146 Ossipov et al.

2001a. Mechanisms of opioid-induced pain and antinoci-ceptive tolerance: descending facilitation and spinaldynorphin. Pain 92:5–9.

Vanderah TW, Suenaga NM, Ossipov MH, Malan TP Jr,Lai J, Porreca F. 2001b. Tonic descending facilitationfrom the rostral ventromedial medulla mediates opioid-induced abnormal pain and antinociceptive tolerance.J Neurosci 21:279–286.

Verge VM, Wiesenfeld-Hallin Z, Hokfelt T. 1993. Chole-cystokinin in mammalian primary sensory neurons andspinal cord: in situ hybridization studies in rat and mon-key. Eur J Neurosci 5:240–250.

Wagner R, Deleo JA. 1996. Pre-emptive dynorphin andN-methyl-D-aspartate glutamate receptor antagonism al-ters spinal immunocytochemistry but not allodynia fol-lowing complete peripheral nerve injury. Neuroscience72:527–534.

Wagner R, DeLeo JA, Coombs DW, Willenbring S, FrommC. 1993. Spinal dynorphin immunoreactivity increasesbilaterally in a neuropathic pain model. Brain Res 629:323–326.

Wang Z, Gardell LR, Ossipov MH, Vanderah TW, BrennanMB, Hochgeschwender U, Hruby VJ, Malan TP Jr, Lai J,Porreca F. 2001. Pronociceptive actions of dynorphinmaintain chronic neuropathic pain. J Neurosci 21:1779–1786.

Waterfield AA, Leslie FM, Lord JA, Ling N, KosterlitzHW. 1979. Opioid activities of fragments of beta-endor-phin and of its leucine65-analogue. Comparison of thebinding properties of methionine- and leucine-enkepha-lin. Eur J Pharmacol 58:11–18.

Waterfield AA, Lord JA, Hughes J, Kosterlitz HW. 1978.Differences in the inhibitory effects of normorphine andopioid peptides on the responses of the vasa deferentia oftwo strains of mice. Eur J Pharmacol 47.

Watkins LR, Kinscheck IB, Kaufman EF, Miller J, Frenk H,Mayer DJ. 1985a. Cholecystokinin antagonists selec-tively potentiate analgesia induced by endogenous opi-ates. Brain Res 327:181–190.

Watkins LR, Kinscheck IB, Mayer DJ. 1985b. Potentiationof morphine analgesia by the cholecystokinin antagonistproglumide. Brain Res 327:169–180.

Way EL, Loh HH, Shen FH. 1969. Simultaneous quantita-tive assessment of morphine tolerance and physical de-pendence. J Pharmacol Exp Ther 167:1–8.

Wegert S, Ossipov MH, Nichols ML, Bian D, VanderahTW, Malan TP Jr, Porreca F. 1997. Differential activitiesof intrathecal MK-801 or morphine to alter responses tothermal and mechanical stimuli in normal or nerve-in-jured rats. Pain 71:57–64.

Welch SP, Singha AK, Dewey WL. 1989. The antinocicep-tion produced by intrathecal morphine, calcium, A23187,U50,488H, [D-Ala2, N-Me-Phe4, Gly-ol]enkephalin and[D-Pen2, D- Pen5]enkephalin after intrathecal adminis-tration of calcitonin gene- related peptide in mice. J Phar-macol Exp Ther 251:1–8.

Wilcox GL. 1991. Excitatory neurotransmitters and pain. In:Bond MR, Woolf CJ, Bond MR, Woolf CJ, editorss.

Proceedings of the VIth World Congress on Pain. Am-sterdam: Elsevier, p 97–117.

Williams JT, Christie MJ, Manzoni O. 2001. Cellular andsynaptic adaptations mediating opioid dependence.Physiol Rev 81:299–343.

Wong CS, Hsu MM, Chou R, Chou YY, Tung CS. 2000.Intrathecal cyclooxygenase inhibitor administration at-tenuates morphine antinociceptive tolerance in rats. Br JAnaesth 85:747–751.

Woolf CJ. 1981. Intrathecal high dose morphine produceshyperalgesia in the rat. Brain Res 209:491–495.

Xu XJ, Wiesenfeld-Hallin Z, Hughes J, Horwell DC, Hok-felt T. 1992. CI988, a selective antagonist of cholecysto-kininB receptors, prevents morphine tolerance in the rat.Br J Pharmacol 105:591–596.

Yaksh TL, Harty GJ. 1988. Pharmacology of the allodyniain rats evoked by high dose intrathecal morphine. J Phar-macol Exp Ther 244:501–507.

Yaksh TL, Harty GJ, Onofrio BM. 1986. High dose ofspinal morphine produce a nonopiate receptor-mediatedhyperesthesia: clinical and theoretic implications. Anes-thesiology 64:590–597.

Yaksh TL, Huang SP, Rudy TA. 1977a. The direct andspecific opiate-like effect of met5-enkephalin and ana-logues on the spinal cord. Neuroscience 2:593–596.

Yaksh TL, Kohl RL, Rudy TA. 1977b. Induction of toler-ance and withdrawal in rats receiving morphine in thespinal subarachnoid space. Eur J Pharmacol 42:275–284.

Yaksh TL, Noueihed R. 1985. The physiology and pharma-cology of spinal opiates. Annu Rev Pharmacol Toxicol25:433–462.

Yaksh TL, Rudy TA. 1976. Analgesia mediated by a directspinal action of narcotics. Science 192:1357–1358.

Yaksh TL, Rudy TA. 1977. Studies on the direct spinalaction of narcotics in the production of analgesia in therat. J Pharmacol Exp Ther 202:411–428.

Yaksh TL, Rudy TA. 1978. Narcotic analgestics: CNS sitesand mechanisms of action as revealed by intracerebralinjection techniques. Pain 4:299–359.

Yaksh TL, Yeung JC, Rudy TA. 1976. Systematic exami-nation in the rat of brain sites sensitive to the directapplication of morphine: observation of differential ef-fects within the periaqueductal gray. Brain Res 114:83–103.

Yeomans DC, Clark FM, Paice JA, Proudfit HK. 1992.Antinociception induced by electrical stimulation of spi-nally projecting noradrenergic neurons in the A7 cate-cholamine cell group of the rat. Pain 48:449–461.

Yeung JC, Rudy TA. 1980a. Multiplicative interaction be-tween narcotic agonisms expressed at spinal and su-praspinal sites of antinociceptive action as revealed byconcurrent intrathecal and intracerebroventricular injec-tions of morphine. J Pharmacol Exp Ther 215:633–642.

Yeung JC, Rudy TA. 1980b. Sites of antinociceptive actionof systemically injected morphine: involvement of su-praspinal loci as revealed by intracerebroventricular in-jection of naloxone. J Pharmacol Exp Ther 215:626–632.

Yeung JC, Yaksh TL, Rudy TA. 1977. Concurrent mapping

Actions of Opioids 147

of brain sites for sensitivity to the direct application ofmorphine and focal electrical stimulation in the produc-tion of antinociception in the rat. Pain 4:23–40.

Zaki PA, Bilsky EJ, Vanderah TW, Lai J, Evans CJ, Porreca F.1996. Opioid receptor types and subtypes: the delta receptoras a model. Annu Rev Pharmacol Toxicol 36:379–401.

Zech DF, Grond S, Lynch J, Hertel D, Lehmann KA. 1995.Validation of World Health Organization Guidelines forcancer pain relief: a 10-year prospective study. Pain 63:65–76.

Zhou Y, Sun YH, Zhang ZW, Han JS. 1992. Acceleratedexpression of cholecystokinin gene in the brain of ratsrendered tolerant to morphine. Neuroreport 3:1121–1123.

Zhou Y, Sun YH, Zhang ZW, Han JS. 1993. Increasedrelease of immunoreactive cholecystokinin octapeptideby morphine and potentiation of mu-opioid analgesia byCCKB receptor antagonist L-365,260 in rat spinal cord.Eur J Pharmacol 234:147–154.

Zhuo M, Gebhart GF. 1992. Characterization of descendingfacilitation and inhibition of spinal nociceptive transmis-sion from the nuclei reticularis gigantocellularis and gi-gantocellularis pars alpha in the rat. J Neurophysiol 67:1599–1614.

Zhuo M, Gebhart GF. 1997. Biphasic modulation of spinalnociceptive transmission from the medullary raphe nucleiin the rat. J Neurophysiol 78:746–758.

148 Ossipov et al.