g lia as th e Òb ad gu ysÓ: im p licatio n s fo r im p ro vin g clin ical …ibg · 2009. 1....

Brain, Behavior, and Immunity 21 (2007) 131–146www.elsevier.com/locate/ybrbi

0889-1591/$ - see front matter 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.bbi.2006.10.011

Norman Cousins Lecture

Glia as the “bad guys”: Implications for improving clinical pain control and the clinical utility of opioids

Linda R. Watkins a,¤,1, Mark R. Hutchinson a, Annemarie Ledeboer a,b, Julie Wieseler-Frank a, Erin D. Milligan a, Steven F. Maier a

a Department of Psychology and the Center for Neuroscience, Muenzinger D-244, Campus Box 345, University of Colorado at Boulder, Boulder, CO 80309-0345, USA

b Avigen, Alameda, CA 94502, USA

Received 16 August 2006; received in revised form 3 October 2006; accepted 4 October 2006Available online 18 December 2006

Abstract

Within the past decade, there has been increasing recognition that glia are far more than simply “housekeepers” for neurons. Thisreview explores two recently recognized roles of glia (microglia and astrocytes) in: (a) creating and maintaining enhanced pain states suchas neuropathic pain, and (b) compromising the eYcacy of morphine and other opioids for pain control. While glia have little-to-no role inpain under basal conditions, pain is ampliWed when glia become activated, inducing the release of proinXammatory products, especiallyproinXammatory cytokines. How glia are triggered to become activated is a key issue, and appears to involve a number of neuron-to-gliasignals including neuronal chemokines, neurotransmitters, and substances released by damaged, dying and dead neurons. In addition, gliabecome increasingly activated in response to repeated administration of opioids. Products of activated glia increase neuronal excitabilityvia numerous mechanisms, including direct receptor-mediated actions, upregulation of excitatory amino acid receptor function, downreg-ulation of GABA receptor function, and so on. These downstream eVects of glial activation amplify pain, suppress acute opioid analgesia,contribute to the apparent loss of opioid analgesia upon repeated opioid administration (tolerance), and contribute to the development ofopioid dependence. The potential implications of such glial regulation of pain and opioid actions are vast, suggestive that targeting gliaand their proinXammatory products may provide a novel and eVective therapy for controlling clinical pain syndromes and increasing theclinical utility of analgesic drugs. 2006 Elsevier Inc. All rights reserved.

Keywords: Interleukin-1; Microglia; Astrocytes; Neuropathic pain; Morphine; Methadone; Analgesia; Tolerance; Dependence; Withdrawal

1. Overview

Within just the past decade, there has been increasing rec-ognition that glia are far more than simply “housekeepers”for neurons, providing neurochemical precursors and energysources to neurons, regulating extracellular ion concentra-tions, removing debris, and so on. It is now known that gliaalso importantly contribute to fever, alterations in sleep, dis-ruption of learning and memory, as well as to neuroinXam-

matory/neurodegenerative conditions such as ischemia/stroke, Alzheimer’s disease, and Parkinson’s disease.

Here the focus is on two recently recognized roles of glia:(a) creating and maintaining pain facilitation, and (b) com-promising the eYcacy of opioids for pain control. Thesepain-relevant topics are of interest primarily because clini-cal pain syndromes occur in epidemic proportions world-wide, including chronic pain states that last for years to alifetime of unremitting pain. In these pain states, hot, coldand hard pressure pains are greatly ampliWed; warm, cooland light touch are now perceived as pain. Such pain ispoorly controlled by currently available therapeutics, allof which were developed to target neurons. Classically,every model used to try to understand such pain focused

* Corresponding author. Fax: +1 303 492 2967.E-mail address: [email protected] (L.R. Watkins).

1 Linda R. Watkins was the recipient of the 2006 Norman CousinsAward.

132 L.R. Watkins et al. / Brain, Behavior, and Immunity 21 (2007) 131–146

exclusively on neurons (Fig. 1). Glia had no role, as theywere not thought of in terms of neural signaling. Despitemany decades of research and drug development, painfacilitation remains unresolved. The discussion to followwill build the case from literature accruing over the past 15years that glia are key players in pain facilitation and a wor-thy target for drug development for clinical pain control.

Regarding the second, inter-related topic of this review,glial regulation of opioid eVects is a new phenomenon thathas begun to develop only within the past 5 years. TheseWndings are intriguing as they shed light on why opioids areexcellent at controlling acute pain but fail to control painfacilitation. The research also reveals that glia importantlycontribute to the development of opioid tolerance, whereby

Fig. 1. Classical and non-classical views of pain transmission and pain modulation (a). Classical pain transmission pathway. When a noxious (painful)stimulus is encountered, such as stepping on a nail as shown, peripheral “pain”-responsive A-delta and C nerve Wbers are excited. These axons relay actionpotentials to the spinal cord dorsal horn. Here, neurotransmitters are released by the sensory neuron and these chemicals bind to and activate postsynapticreceptors on pain transmission neurons (PTNs) whose cell bodies reside in the dorsal horn. Axons of the PTNs then ascend to the brain, carrying informa-tion about the noxious event to higher centers. The synapse interconnecting the peripheral sensory neuron and the dorsal horn PTN is shown in detail in(b) and (c). (b): Normal pain. Under basal conditions, pain is not modulated by glia. Under these circumstances, glia are quiescent, and thus not releasingpain modulatory levels of neuroexcitatory substances. Information about noxious stimuli arrives from the periphery along A-delta and C Wbers, causingthe release of substance P and excitatory amino acids (EAAs) in amounts appropriate to the intensity and duration of the initiating noxious stimulus. Acti-vation of neurokinin-1 (NK-1) receptors by substance P and activation of AMPA receptors by EAAs cause transient depolarization of the PTNs, therebygenerating action potentials that are relayed to brain. NMDA-linked channels are silent as they are chronically “plugged” by magnesium ions. (c) Painfacilitation: classical view. In response to intense and/or prolonged barrages of incoming “pain” signals, the PTNs become sensitized and over-respond tosubsequent incoming signals The intense and/or prolonged barrage depolarizes the PTNs such that the magnesium ions exit the NMDA-linked channel.The resultant inXux of calcium ion activates constitutively expressed nitric oxide synthase (cNOS), causing conversion of L-arginine to nitric oxide (NO).Because it is a gas, NO rapidly diVuses out of the PTNs. This NO acts presynaptically to cause exaggerated release of substance P and EAAs. Postsynapti-cally, NO causes the PTNs to become hyperexcitable. Glia have not been considered to have a role in creating pain facilitation in this neuronally drivenmodel. (d): Pain facilitation: new view. Here, glial activation is conceptualized as a driving force for creating and maintaining pain facilitation. The role ofglia is superimposed on the NMDA-NO-driven neuronal changes detailed in (c), so only the aspects added by including glia in the model are describedhere. Glia are activated (shown as hypertrophied relative to (b), as this reXects the remarkable anatomical changes that these cells undergo on activation)by three sources: bacteria and viruses which bind speciWc activation receptors expressed by microglia and astrocytes; substance P, EAAs, fractalkine, andATP released by A-delta and/or C Wber presynaptic terminals (shown here) or by brain-to-spinal cord pain enhancement pathways (not shown); and NO,prostaglandins (PGs) and fractalkine released from PTNs. Following activation, microglia and astrocytes cause PTN hyperexcitability and the exagger-ated release of substance P and EAAs from presynaptic terminals. These changes are created by the glial release of NO, EAAs, reactive oxygen species(ROS), PGs, proinXammatory cytokines (for example, IL1, IL6 or TNF), and nerve growth factor. ModiWed by Journal of Internal Medicine with permis-sion, from Trends in Neuroscience.

L.R. Watkins et al. / Brain, Behavior, and Immunity 21 (2007) 131–146 133

opioids become progressively less eVective in pain control,requiring progressively more opioid to suppress pain.Lastly, the newest wrinkle in this research will be described,that being the involvement of glia in the development ofmorphine dependence. As a consequence of repeated opioiduse, animals become dependent on opioids such that whenopioid dosing is stopped, animals (including humans) entera state of opioid withdrawal. As will be described, glia arenow also implicated in opioid dependence/withdrawal.

Before proceeding, terminology relevant to this reviewwill be brieXy introduced. “Pain” refers to the unpleasantexperience associated with actual or perceived tissue dam-age by a noxious (damaging) stimulus. Pain, as an experi-ence, not only includes the perception of a simple sensorystimulus but also the cognitive analysis and emotionalresponses associated with that. Quantifying the pain expe-rience in laboratory animals is arguably impossible. Giventhat, terminology was created that refers to the “lowerorder” quantiWable aspects of pain. “Nociception” (“noci”meaning “pain”) is used by the Weld to describe neural,physiological or behavioral responses to stimuli thathumans would typically report as painful. Nociception isfurther broken down to reXect responding to noxious stim-uli (pinch, for example) as if they are more intense thannormal (hyperalgesia) versus responding to normallyinnocuous stimuli (light touch, for example) as if they werenoxious (allodynia). For the sake of ease of communica-tion in this general review, however, the term “pain” willrefer to both animal and human responses and “pain facil-itation” or “enhanced pain”, for example, will be usedrather than more specialized terminology.

2. Glia

The work to be reviewed focuses on two types of glialcells, astrocytes and microglia. When these cells becomeactivated, they (a) upregulate cell-type speciWc activationmarkers (allowing this reXection of activation to be easilyvisualized using immunohistochemistry) and (b) release avariety of substances (e.g., proinXammatory cytokines, che-mokines, ATP, excitatory amino acids, nitric oxide, etc.) thatenhance pain by increasing the excitability of nearby neu-rons. Microglia and astrocytes are not likely to be the onlynon-neuronal cells involved in pain enhancement. Endothe-lial cells, Wbroblasts, oligodendroglia, and other cell types inboth the spinal cord and overlying meninges can also pro-duce many of the same neuroexcitatory substances as doastrocytes and microglia. However, study of the involvementof such cells has been hindered by the lack of their expres-sion of upregulatable activation markers. As microglia andastrocytes have proven to be the cell types most amenable tostudy, virtually all studies to date have focused on them.

2.1. Microglia

Microglia constitute 5–12% of all cells in the CNS and5–10% of all glia. Under basal conditions, microglia per-

form immune surveillance (Raivich, 2005). A shift to anactivated state can occur rapidly as microglia are exquisitesensors of changes in their microenvironment. Stimuli thatcan activate microglia include trauma, infection, ischemiaand neurodegeneration. Activation results in changes inmorphology (e.g., retracted processes), proliferation, upreg-ulated receptor expression (e.g., complement receptors,scavenger receptors), and changes in function (e.g., migra-tion to sites of damage, phagocytosis, release of proinXam-matory mediators). After resolution of challenge, microgliacan either return to their basal state or may enter a“primed” state for prolonged periods of time (Perry et al.,1985). “Primed” microglia do not actively produce proin-Xammatory cytokines (and indeed may instead release anti-inXammatory cytokines) but now over-respond to newchallenges, both in the speed and magnitude of release ofproinXammatory products (Perry et al., 1985), similar inspeed and magnitude to a secondary peripheral immuneresponse. Microglial priming may prove to be important inexplaining why some people greatly over-respond, inintensity and duration, to a pain-evoking event. Evidence todate from animal models suggests that prior microglialactivation can leave these cells primed, such that a laterpain-evoking stimulus produces both enhanced spinal pro-inXammatory cytokine production and enhanced pain(Spataro et al., 2006).

2.2. Astrocytes

Astrocytes constitute 40–50% of all glial cells (Aldsko-gius and Kozlova, 1998) and outnumber neurons (Leeet al., 2000). Astrocytes tightly enwrap the vast majority ofsynapses in the CNS and actively modulate neuron-to-neuron synaptic communication. The dynamic astrocyteregulation of synaptic communication has given rise to theterm “tripartite synapse” to emphasize that astrocytes, inaddition to the neuronal presynaptic and neuronal postsyn-aptic processes, are integral parts of neuronal signaling(Haydon and Carmignoto, 2006). Indeed, astrocytes areimportant contributors to synaptic “memory”, as prior syn-aptic activity leads, at later times, to greater astrocyteresponses to subsequent synaptic input (Pasti et al., 2001).Beyond the synapse, astrocytes also are in intimate contactwith neuronal cell bodies, dendrites, and nodes of Ranvier.They extend perivascular endfeet to contact capillaries andform a border layer in the meninges (Aldskogius andKozlova, 1998). Under basal conditions, astrocytes provideneurons with energy sources and neurotransmitter precur-sors (Tsacopoulos, 2002), provide trophic support, regu-late extracellular ions and neurotransmitters, and regulateneuronal survival and diVerentiation, neurite outgrowth,and formation of synapses (Perea and Araque, 2002).Astrocytes are activated by trauma, inXammation or infec-tion, with progressive upregulation in glial Wbrillary acidicprotein (GFAP) expression, morphology and proliferationupon persistent activation (Aldskogius and Kozlova,1998).


2.3. Microglia–astrocyte interactions

In vivo, astrocytes and microglia interact. Their releasedproducts can synergize, and substances released by one canactivate the other. Regarding synergy, proinXammatorycytokines can synergize with each other as well as with neu-rotransmitters and neuromodulators, such as norepineph-rine, prostaglandin E2 (PGE2), and nitric oxide (for review,see (Watkins et al., 2006)). Pain relevant examples includethe synergy of TNF and IL-1 with ATP to enhance PGE2release (Loredo and Benton, 1998); nitric oxide potentiat-ing IL-1 induced PGE2 production and substance P releasefrom sensory aVerent terminals in spinal cord (Moriokaet al., 2002); and substance P potentiating IL-1 inducedrelease of IL-6 and PGE2 from human spinal cord astro-cytes (Palma et al., 1997). Regarding cross-stimulationbetween glia: (a) astrocytes release substances that stimu-late microglial activation, proliferation, and production ofnitric oxide; and (b) microglia release substances thatinduce astrocyte activation, expression of adhesion mole-cules, functioning as antigen presenting cells, and release ofglutamate, TNF, IL-1 and nitric oxide (for review, see(Watkins et al., 2006)). These microglia–astrocyte interac-tions are consistent with the developing pain literature thatsuggests that microglia are the Wrst glial type to becomeactivated and that their activation leads in turn to therecruitment of nearby astrocytes, such that both cell typescan contribute to observed downstream alterations in pain(Ledeboer et al., 2005; TawWk et al., 2006).

3. Glial dysregulation of pain

Under normal situations, when spinal glia are in theirbasal state, they have little-to-no inXuence on pain respon-sivity (Fig. 1). Various drugs that suppress glial functionhave consistently failed to alter responsivity to heat ormechanical stimuli in normal animals (Hashizume et al.,2000; Ledeboer et al., 2005; Meller et al., 1994). While datafrom knockouts reveal reductions in basal pain respon-sivity in adult mice with disrupted IL-1 signaling, thesechanges appear related to developmental eVects as the sameeVects occur when rats pups are treated only in utero withIL-1ra (Wolf et al., 2003). Thus, the weight of evidence sup-ports that when astrocytes and microglia are not in theiractivated state, they do not appear to be important regula-tors of pain transmission. Having said this, basal cytokinelevels are integral to maintaining neuronal plasticity.

When glia become activated, pain is now aVected. Onesituation where spinal glia can become activated is inresponse to peripheral immune challenges that activateimmune-to-brain communication (Dantzer, 2004; Maierand Watkins, 1998). Peripheral immune activated signalingto the CNS induces sickness responses. Work from our lab-oratory has documented that sickness responses includeenhanced pain responsivity, in addition to classically recog-nized fever, increases in sleep, decreases in food and waterintake, and generalized suppression of behavior (Dantzer,

2004; Maier and Watkins, 1998). Studies of brain-mediatedsickness responses revealed that glial activation and conse-quent proinXammatory cytokine release are key steps in thegeneration of these survival-oriented changes in CNS func-tion. These early studies predicted that sickness-inducedpain enhancement would likewise be mediated by spinalglial activation and proinXammatory cytokines. Thisindeed turned out to be true (Watkins and Maier, 2000).

It is now known that glial regulation of pain extends wellbeyond enhanced pain responses that occur as a naturalcomponent of the sickness response. Both astrocytes andmicroglia in spinal cord are activated (as inferred from glialactivation markers) in response to inXammation or damageto peripheral tissues, peripheral nerves, spinal nerves, orspinal cord (McMahon et al., 2005; Watkins and Maier,2003). The generally accepted view has developed thatmicroglia are activated Wrst, followed by astrocytes (Tangaet al., 2004). While microglia were thought to fade, andastrocytes to increase, in prominence over time (Ledeboeret al., 2005; Raghavendra et al., 2003), a prolonged role formicroglia has recently been proposed (TawWk et al., 2006).

Regarding the mechanisms by which glial activationinXuences pain, work from our laboratory and many othershave revealed that spinal glial activation and proinXamma-tory cytokines have been implicated in enhanced pain asso-ciated with every animal model examined to date (Table 1),with rare exception (Ledeboer et al., 2006a). This conclu-sion is based on the facts that intrathecal (into the cerebro-spinal Xuid [CSF] space surrounding the spinal cord)injection of proinXammatory cytokines enhances pain (Fal-chi et al., 2001; Reeve et al., 2000; Tadano et al., 1999), andglial and cytokine inhibitors prevent and/or reverse suchpain enhancements (McMahon et al., 2005; Watkins andMaier, 2003), as does knocking out IL-1 signaling (Honoreet al., 2006; Wolf et al., 2006).

Importantly, we have shown that ongoing inhibition ofspinal cord proinXammatory cytokines can reverse estab-lished nerve injury-induced pain facilitation (i.e., neuro-pathic pain): (a) for 3+ months (Milligan et al., 2006),suggestive that other mechanisms will not reinstate painfacilitation in the absence of these cytokines, and (b) evenafter pain facilitation has been maintained continuously for1–2 months (Milligan et al., 2006; TawWk et al., 2006). Ourobservations provide strong support for the conclusion thatglia and proinXammatory cytokines do not just induce painfacilitation but, rather, are key mediators of pain mainte-nance as well. This is especially important when consideringthe potential clinical utility of targeting proinXammatorycytokines for pain control as pain syndromes in people,such as neuropathic pain, will typically be long establishedand require long-term treatment. Indeed, in keeping withthe recognition that the relative balance between pro- andanti-inXammatory cytokine levels may be key to the eVectsobserved, increases in proinXammatory cytokines and/ordecreases in anti-inXammatory cytokines have been foundin spinal CSF samples from chronic pain patients diag-nosed with complex regional pain syndrome, Wbromyalgia


and neuropathic pain (Alexander et al., 2005; Backonjaet al., 2006).

4. How do spinal glia “know” to become activated?

A key question is: “What is released in spinal cord, as aresult of inXammation/trauma in the periphery that “tells”glia to activate?” Increasing focus is being directed atmicroglia as the “trigger”, as the initiator of pain facilita-tion. Microglia are exquisitely sensitive to signals of “notself” or “not normal”, leading them to be far more reactivein responding to CNS challenges than other candidate celltypes (Kreutzberg, 1996). Similarly, spinal cord microgliaare the earliest glial cell type activated in response toperipheral inXammation and injury, based both on the abil-ity of the microglial inhibitor minocycline to prevent theonset of enhanced pain and the rapid upregulation ofmicroglial activation markers relative to those of astrocytes(Ledeboer et al., 2005; Raghavendra et al., 2003). It appears

that this initial microglial activation then recruits the acti-vation of astrocytes (Tanga et al., 2004).

A number of putative neuron-to-glia signals have nowbeen documented (Fig. 2). These are reviewed below.

4.1. Neuronal chemokines

Chemokines are chemo attractant cyto kines, consistingof a family of 50+ members. These chemokines are groupedinto subfamilies, based on the number and positioning ofamino acid(s) (X) inserted between cysteine (C) residues:CC, CXC, XC, and CXXXC (CX3C). While initiallythought to be of immune origin, a number of these proteinscan be produced and released by neurons as well (Stievanoet al., 2004).

Fractalkine was the Wrst chemokine discovered to be aneuron-to-glia signal. Fractalkine is also known as thechemokine ligand (L), CX3CL1. It is the only knownmember of this chemokine subfamily and the only known

Table 1Summary of pain models with glial/cytokine involvement

Abbreviations: IL-1ra, interleukin-1 receptor antagonist; sTNFR, soluble TNF receptor; IL-10, interleukin-10.ModiWed and updated from Ledeboer et al. (Ledeboer et al., 2006a).For complete references, see (Hains and Waxman, 2006; Honore et al., 2006; Ledeboer et al., 2006a; Obata et al., 2006; Watkins et al., 2005).

A. Glial activation spinal cord (microglia and/or astroglia): pain models associated with upregulation of spinal cord microglial and/or astrocyte activation markers

Complete Freund’s adjuvant, subcutaneousFormalin, subcutaneousPhospholipase A2, subcutaneousZymosan, subcutaneousSciatic nerve injury (chronic constriction injury)Partial sciatic nerve ligationSciatic nerve inXammation with zymosanSciatic nerve inXammation with HIV-1 gp120Sciatic nerve inXammation with phospholipase A2Spinal nerve transectionSpinal nerve root injurySpinal cord injuryBone cancerHIV-1 gp120, intrathecalChronic opioids; opioid withdrawal-induced hyperalgesia

Model Intervention

B. Prevention or reversal of pain by inhibition of spinal glial activation or proniflammatory cytokine actionsCarrageenan, subcutaneous Minocycline, IL-1 knockoutComplete Freund’s adjuvant, subcutaneous IL-1ra, IL-1 knockoutFormalin, subcutaneous Fluorocitrate, IL-1ra, minocycline; IL-1 knockoutPhospholipase A2, subcutaneous Fluorocitrate, IL-1ra, sTNFRZymosan, subcutaneous Fluorocitrate;Sciatic nerve injury (chronic constriction injury) IL-1ra, IL-10; IL-1 knockoutSciatic nerve inXammation with zymosan Fluorocitrate, minocycline

IL-1ra, sTNFR, IL-6 antibodySciatic nerve inXammation with phospholipase A2 IL-1ra, anti-IL-6, IL-10Sciatic nerve tetanic stimulation FluorocitrateSpinal nerve transection Propentofylline, minocycline,

IL-1ra, sTNFR, anti-IL-6,Spinal nerve root injury Methotrexate; IL-1 knockoutSpinal cord injury IL-10, IL-1ra, minocyclineHIV-1 gp120, intrathecal Fluorocitrate, IL-1ra, sTNFR,

MinocyclineDynorphin, intrathecal IL-1ra, IL-10Fractalkine, intrathecal Minocycline, IL-1ra, anti-IL-6


ligand of the fractalkine receptor (R), CX3CR1. This is anunusual proWle for chemokines, as typically a single che-mokine binds to multiple chemokine receptors and a sin-gle chemokine receptor binds to multiple chemokines (forexcellent summary tables, see http://www.nature.com/bjp/journal/vgrac/ncurrent/full/0706562a.html). An additionalunique feature of fractalkine is that it is tethered to theextracellular membrane of neurons. This attachment iscleaved upon strong neuronal activation, thereby forminga diVusible signaling molecule (Chapman et al., 2000). Ourinvestigations of the distribution of fractalkine and itsreceptor in spinal cord revealed an intriguing pattern:fractalkine was expressed only by neurons and its receptorwas expressed only by microglia (Verge et al., 2004). Weobserved this arrangement under both basal conditionsand in response to either inXammation or trauma to thesciatic nerve (Verge et al., 2004). Intriguingly, the fractal-kine receptor is upregulated in pain modulatory regionsof spinal cord under neuropathic pain conditions (Lindiaet al., 2005; Verge et al., 2004) and in response to anklemonoarthritis (Shan et al., 2006). Taken together, suchanatomical data suggested that this chemokine mightserve as a neuron-to-microglia signal, triggering microg-lial activation. The only known exceptions to this ana-tomical distribution are multiple sclerosis inXammatorylesions (Sunnemark et al., 2005) and spinal nerve trauma(Lindia et al., 2005) where novel astrocyte expression offractalkine has been observed (Fig. 3).

Upon intrathecal injection of fractalkine, we observedenhanced pain responsivity (Milligan et al., 2005; Milliganet al., 2004). We found that this pain enhancement by frac-talkine is mediated via CX3CR1 as neutralizing antibodiesagainst this receptor prevent pain enhancement by intrathe-cal fractalkine (Milligan et al., 2005). As predicted bymicroglial expression of CX3CR1 in spinal cord (Vergeet al., 2004), our studies revealed that fractalkine enhancespain responsivity via microglial activation, as painenhancement is blocked by the microglial inhibitor minocy-cline. Furthermore, fractalkine enhances pain via therelease of spinal proinXammatory cytokines and nitricoxide (Milligan et al., 2005). Electrophysiological studiesreveal that fractalkine causes hyper-responsivity of spinalneurons to brush and pinch, as well as increases in the num-bers of neurons exhibiting prolonged afterdischarges indic-ative of spontaneous pain and central sensitization(Owolabi and Saab, 2006). Fractalkine binding to its recep-tor leads to downstream changes previously implicated inpain enhancement. For example, fractalkine binding leadsto NF!B and p38 MAP kinase activation (Stievano et al.,2004) and production of proinXammatory cytokines andchemokines (Johnston et al., 2004; Stievano et al., 2004).

Importantly, endogenous fractalkine modulates pain aswell. Blocking spinal fractalkine receptors delays and/orreverses neuropathic pain arising from sciatic nerve inXam-mation, sciatic nerve trauma, or ankle monoarthritis (Milli-gan et al., 2004; Shan et al., 2006), supporting the idea thatendogenous fractalkine shed from neurons is importantlyinvolved in the initiation and maintenance of neuropathicpain.

Beyond fractalkine, other neuronally derived chemo-kines may be neuron-to-glia signals. Two chemokineligands of CXCR3 (CCL21/secondary lymphoid tissue che-mokines; CXCL10/interferon-inducible protein of 10 kDa[IP-10]) are rapidly induced in, and released from, damagedneurons (Biber et al., 2002; Rappert et al., 2004). Neuronsnot only produce CXCR3 ligands, but also package them invesicles and transport them from the primary lesion site todistant presynaptic terminals. This allows for remote acti-vation of glia, as occurs in spinal cord following peripheralnerve injury. Astrocytes and microglia each expressCXCR3 mRNA and protein (Biber et al., 2002). In addi-tion, stimulation of glia with CXCR3 agonists induce cal-cium and chloride transients as well as chemotaxis,indicating activation (Biber et al., 2002; Rappert et al.,2002). The importance of this communication pathway issupported by the fact that mice deWcient for CXCR3 dis-play no secondary activation of microglia in response toentorhinal cortex lesions. Whether CXCR3 agonists inducepain facilitation upon intrathecal injection has not yet beeninvestigated.

A last neuronally derived chemokine that is intriguingwith regards to pain-related neuron-to-glia signaling isCCL2, also known as monocyte chemoattractant protein-1(MCP-1). CCL2/MCP-1 and its cognate receptor CCR2 areupregulated in damaged dorsal root ganglion neurons

Fig. 2. Summary of neuronal signals binding to microglial receptors lead-ing to microglial activation and associated with pain facilitation. Neuronsrelease these signals following speciWc stimulation or as a consequence ofneuronal damage. As described in the main text, these receptors and sig-nals have been associated with chronic pain states. Receptors are pre-sented in bold. Neuronal signals bind chemokine receptors (Fractalkine[FKN] binds CX3CR1; CCL2 binds CCR2; IP-10 [interferon-inducibleprotein] and CCL21 [secondary lymphoid tissue chemokines] bindCXCR3); toll-like receptors (high mobility group box 1 [HMGB1] bindsToll Like Receptors 2 and 4 [TLR2 and TLR 4] and proteoglycans, heatshock proteins [HSPs], saturated fatty acids, and cations bind TLR4); andlysophosphatidic acid binds LPA1. Classical pain neurotransmitters arereleased by neurons and act on speciWc microglial receptors such as aden-osine triphosphate (ATP) binding P2X4; Substance P binding NK-1; andexcitatory amino acids [EAA] binding NMDA and AMPA receptors.

http://www.nature.com/bjp/journal/vgrac/ncurrent/full/0706562a.html




(White et al., 2005). While CCR2/MCP-1 is absent in nor-mal CNS, levels rise following peripheral nerve injury asMCP-1 produced by injured neurons of the dorsal rootganglia is transported to sensory terminals in the spinalcord to form a releasable pool (Zhang and De Koninck,2006). These sensory projections to the spinal cord appearto be the predominant source of CCL2/MCP-1 followingperipheral nerve injury (Zhang and De Koninck, 2006). Inaddition, CCR2 is upregulated in spinal cord microglia andastrocytes following peripheral nerve injury, and theseappear to be the only cell types expressing CCR2 in spinalcord (Abbadie et al., 2003; El-Hage et al., 2006). NeuronalCCR2/MCP-1 induction is followed by surroundingmicroglial activation, suggestive that CCR2/MCP-1 mayserve as a trigger for spinal microglial activation (Zhangand De Koninck, 2006). Intrathecal injection of CCR2/MCP-1 produces pain facilitation (Tanaka et al., 2004) andintrathecal administration of neutralizing antibodies to

CCR2/MCP-1 suppresses neuropathic pain (Abbadie et al.,2003). CCR2 knock out mice exhibit less activation of spi-nal cord astrocytes (as reXected by GFAP expression) andless activation of spinal cord microglia (as reXected byphospho-p38 MAP kinase expression) after peripheralnerve injury (Abbadie et al., 2003). In turn, CCR2 knockout mice exhibit greatly reduced central sensitization toinXammatory stimuli and are reported to fail to exhibitneuropathic pain (Abbadie et al., 2003).

4.2. Neurotransmitters

Sensory neurons relay messages of “pain” to the spinalcord using a variety of neurotransmitters, including sub-stance P, excitatory amino acids, and ATP. As such, eachmay potentially be involved in glial activation. Spinal cordastrocytes express receptors for substance P and, uponstimulation with substance P, these astrocytes release PGE2

Fig. 3. (A) The classical view of opioid analgesia proposes (¡)-opioid agonist isomers stereoselectively bind to classical opioid receptors producing aninhibitory inXuence on nociceptive signal transmission. (B) A growing body of literature suggests that the classical view of analgesia ignores an importantnociceptive modulatory inXuence driven by opioid induced glial activation resulting from opioid agonists binding to glial opioid binding receptors whichin turn increase proinXammatory cytokine (such as interleukin-1) expression and release leading to a decrease in opioid eYcacy at the neuronal compo-nent. (C) (¡)-Opioid antagonists bind to both the neuronal and glial components resulting in blockade of any potential opioid analgesia and glial activa-tion. (D) Due to the stereoselectivity of neuronal opioid receptors only the (¡)-isomer of opioid agonists and antagonists are able to bind. Therefore, whena combination of an opioid (+)-antagonist and an opioid (¡)-agonist are introduced to this system, the (+)-antagonist is unable to bind to the stereoselec-tive neuronal opioid receptor, but is able to block the non-stereoselective glial site. The opioid (¡)-agonist can act freely at the neuronal opioid receptor,but is unable to bind to the glial site due to its blockade by the (+)-antagonist. Therefore, this situation produces opioid receptor mediated analgesia with-out the opposing force of opioid induced glial activation, thereby potentiating opioid analgesia. (E) Based on the above hypothesized series of events thisschematic diagram demonstrates how analgesia decreases over time as a direct result of increased glial activation.


and IL-6 (Marriott et al., 1991). Spinal cord astrocytes alsoexpress metabotropic glutamate receptors, ionotropic non-NMDA receptors (AMPA and kainate) and NMDAreceptors (Aicher et al., 1997; Besong et al., 2002). NMDAreceptor activation leads to the activation of spinal microg-lia and release of both IL-1 and nitric oxide (Tikka andKoistinaho, 2001).

Of the neurotransmitters relaying “pain”, ATP hasrecently become the prime focus of studies exploring neu-ron-to-glia signaling. There is now strong evidence that theexpression of a speciWc subset of spinal cord ATP receptors,termed P2X4, is normally low under basal conditions.Intriguingly, P2X4 receptor expression strongly upregu-lates in response to peripheral nerve injury, but only onmicroglia (Inoue, 2006). The timecourse of P2X4 upregula-tion mirrors the development of neuropathic pain. Suppres-sion of P2X4 receptor signaling delays and reducesneuropathic pain (Inoue, 2006). Indeed, intrathecal injec-tion of microglia activated by ATP in vitro proved suYcientto enhance pain (Tsuda et al., 2003). Similar to fractalkine,ATP activates microglial p38 MAP kinase and leads to therelease of microglial proinXammatory cytokines (Inoue,2006) and plasminogen, a protein which enhances NMDAreceptor function.

In addition, under conditions that induce pain facilita-tion, neurons release neuromodulatory substances includ-ing nitric oxide and PGE2 and enhance the production ofthe neurotransmitter dynorphin. We have recently reportedthat neuronal nitric oxide enhances spinal production andrelease of TNF, IL-1 and IL-6 (Holguin et al., 2004), andPGE2 has been reported to be important in the initiation ofboth glial activation and neuropathic pain, but not in themaintenance of these eVects (Takeda et al., 2005). Lastly,the pain enhancing eVects of spinal dynorphin haverecently been linked to selective activation of microglial p38MAP kinase which, in turn, leads to the release of PGE2and IL-1 (Laughlin et al., 2000; Svensson et al., 2005).

4.3. Substances released by damaged, dying and dead neurons

Substances released by neurons in response to their dam-age or death are likely relevant as neuron-to-glia signals, asdegeneration of sensory axons occurs within the spinal cordafter peripheral nerve injury, dorsal root ganglion trauma,or spinal nerve injury (Costigan et al., 1998). As the clear-ance of myelin and other cellular debris is a remarkablyprotracted process continuing for many years in the humanCNS compared to the periphery (Vargas et al., 2006), thisimplies that substances released by injured neurons undersuch conditions may be able to provide ongoing stimula-tion to maintain equally protracted glial activation. Inaddition, there are reports of neurons in dorsal spinal corddying secondary to peripheral nerve injury (Scholz et al.,2005). ATP is one major substance released by damagedand dying cells. As it is also released as a neurotransmitterin response for painful stimuli, its involvement in glial acti-

vation has already been reviewed, above. Other candidatesubstances, by and large, represent ligands for scavengingtoll-like receptors (TLRs). The TLR family of receptors areexpressed primarily by immune and immune-like cells thatevolved to detect “danger signals” such as conserved motifson bacteria and viruses, cell membrane and nuclear compo-nents released by damage, chaperone proteins (e.g., heatshock proteins; HSP) released by damage, etc. In responseto nerve injury, TLR1, TLR2, and TLR4 have each beenreported to be upregulated in CNS, linked to production ofproinXammatory cytokines (e.g., TNF, IL-1) and chemo-kines (e.g., CCL2/MCP-1) (Owens et al., 2005).

In addition, lysophosphatidic acid (LPA) is a membranephospholipid released in response to neuronal damage,which acts via the LPA(1) receptor rather than via TLRs.LPA may be an additional neuron-to-glia signal linked topain enhancement (Inoue et al., 2004). LPA activates astro-cytes, leading to the expression of IL-1, IL-6 and nervegrowth factor, an additional glial product linked to painfacilitation (Watkins et al., 1997). Indeed, intrathecal LPApersistently enhances pain responses whereas intrathecalanti-sense oligodeoxynucleotides (ODN) to the LPA recep-tor inhibited peripheral nerve injury induced neuropathicpain (Inoue et al., 2004).

High Mobility Group Box 1 (HMGB1) is an example ofa nuclear protein which, upon extracellular release by celldamage or death, is a potent inXammatory mediator (Har-ris and Andersson, 2004). Extracellular HMGB1 binds toand activates TLRs (Park et al., 2006). In keeping with itspotential role as a neuron-to-glia signal, we have shownthat HMGB1 enhances pain responsivity upon intrathecaladministration (O’Connor et al., 2003).

The last candidate of this group is a TLR receptor(TLR4) rather than a ligand per se. TLR4 knockout miceand rats intrathecally administered TLR4 antisense ODNeach display reduced: neuropathic pain, spinal microglialactivation, and spinal proinXammatory cytokines. TLR4 isexpressed only by microglia (Lehnardt et al., 2002) and isupregulated in microglia in response to spinal nerve injurycorrelated with enhanced pain (Tanga et al., 2004). Whilelipopolysaccharide (a component of bacterial cell walls) iswell known as a TLR4 agonist, other candidate signalsinclude saturated fatty acids (Hwang, 2001), ATP, heatshock proteins, and cations (Tanga et al., 2005). Thus, whilethe speciWc ligand(s) responsible for enhanced pain result-ing from TLR4 activation has yet to be identiWed, theinvolvement of TLR4 remains intriguing nonetheless.

4.4. Taking this all together

A number of glial activators have been discussed above,based on studies of such mediators in isolation. It isunlikely that this accurately reXects what happens in vivo.In vivo, it makes little sense that any of these mediators actin isolation. Rather, it is far more likely that glia shoulddetect patterns. Just as various proinXammatory substancescan act in synergy to stimulate glia, so logically should


putative neuron-to-glia signals. We expect that glia shouldbe far more readily triggered to activate in response to aconstellation of substances that, taken together, relay themessage of peripheral nerve or tissue injury. Indeed, earlyevidence from our laboratory suggests this will be the case(Wieseler-Frank et al., 2004). The existence of multiple sig-nals described above suggests that there could be patternsof signals to glia that indicate to glia the type of cell dam-aged, the nature of the damage, and the severity of damage.

5. How can glial activation aVect neuronal activity?

It is remarkable that diverse enhanced pain states are allsuppressed by inhibiting glial function (Watkins and Maier,2003). However, this does not provide insight into what theglial products are that enhance pain. This issue is, in part,relatively simple and, in part, complicated. “Relatively sim-ple” refers to situations in which pain enhancement can beclearly linked to the release of classical glial products, suchas proinXammatory cytokines. “Relatively” simple must bekept in mind however as sensory neurons have been shownto produce and transport TNF to their presynaptic termi-nals in spinal cord after peripheral nerve injury (Shubayevand Myers, 2002). Whether TNF is released from thesedamaged axons into spinal cord is as yet unknown, butwould appear reasonable to assume. Thus, the picture evenfor proinXammatory cytokines may not be as clear as itbegan. The “in part, complicated” refers to the fact that fewsubstances are produced only or predominantly by glia.Neurons, like glia, can produce nitric oxide, ATP, prosta-glandins, excitatory amino acids, and so on, blurring con-clusions whether blockade of enhanced pain byantagonists/inhibitors to such mediators reXect actions onneurons or glia.

Thus, actions of proinXammatory cytokines on neuronswill be described here, as it is the clearest example of howglial products can aVect neuronal activity. This is not nearlyas simple as merely binding of proinXammatory cytokinesto known neuronal receptors (Dame and Juul, 2000;Holmes et al., 2004; Ohtori et al., 2004) so to induce rapidincreases of neuronal excitability (Oka et al., 1994; Reeveet al., 2000; Samad et al., 2004). In addition, IL-1 has beenshown, including in spinal dorsal horn, to enhance neuroex-citability via indirect actions; namely, via increasing cal-cium conductivity of neuronal NMDA receptors (Samadet al., 2004; Viviani et al., 2003). This occurs via intracellu-lar signaling pathways leading to the phosphorylation ofNMDA receptor subunits. TNF increases the conductivityof glutamatergic AMPA receptors as well (De et al., 2003)and increases spontaneous neurotransmitter release frompresynaptic terminals (Grassi et al., 1994). TNF, and to alesser extent, IL-1, upregulate the neuronal cell surfaceexpression of both AMPA and NMDA receptors whiledownregulating cell surface expression of receptors for theinhibitory neurotransmitter, GABA; hence, creating anoverall increase in neuronal excitatory tone (Stellwagenet al., 2005). Lastly, we have shown that IL-1 is implicated

in neuropathy induced downregulation of G protein-cou-pled receptor kinase 2 (GRK2) expression in dorsal hornneurons, an action predicted to increase neuronal excitabil-ity by decreasing receptor desensitization (Kleibeuker et al.,2006).

Beyond these actions, proinXammatory cytokines alsolead to the release of a host of neuroexcitatory substances,including more proinXammatory cytokines, nitric oxide,ATP, prostaglandins, nerve growth factors, reactive oxygenspecies, proinXammatory chemokines (e.g., CCL2/MCP-1,CXCL8/IL-8, CXCL10/IP-10) and BDNF (Inoue, 2006;John et al., 2005; Sperlagh et al., 2004; Watkins et al., 1999).They can also indirectly lead to elevations in extracellularglutamate levels via downregulation of glial and neuronalglutamate transporters that serve to keep extracellular glu-tamate levels low under normal conditions (TawWk et al.,2006). Thus, taken together, proinXammatory cytokinesexert multiple eVects with the end result of neuroexcitation.

6. Non-neuronal cells beyond glia: might they also contribute to pain enhancement?

Before leaving the topic of pain facilitation, it is worthnoting that astrocytes and microglia may well not be theonly non-neuronal cell types capable of enhancing pain inspinal cord. As noted above (Section 2), other cell types inspinal cord can generate proinXammatory cytokines andother neuroexcitatory substances, such as Wbroblasts, peri-vascular macrophages, dendritic cells, and endothelial cells.The potential involvement of such cells in pain facilitationhas not yet been explored.

Intriguingly, immunocompetent cells exist outside thespinal cord parenchyma as well; that is, in the meningesthat enwrap and protect the spinal cord, enclosing the CSFspace. Meningeal tissues include many cell types capable ofproducing proinXammatory substances, including Wbro-blasts, dendritic cells, macrophages, endothelial cells, andmast cells (Wieseler-Frank et al., 2006a). Indeed, we havenow demonstrated that the spinal meninges elevate TNF,IL-1, IL-6 and iNOS gene expression and release TNF, IL-1and IL-6 in response to immune activators such as HIV-1gp120 administered intrathecally at a dose that enhancespain (Wieseler-Frank et al., 2006b). As we have replicatedthese eVects in vitro in the absence of circulating whiteblood cells, these proinXammatory responses are being gen-erated by cells intrinsic to the meninges rather than byimmune cells recruited from the blood (Wieseler-Franket al., 2006a). Intriguingly, similar upregulation of proin-Xammatory cytokines in the meninges occur followingperipheral nerve injury, implying that the meninges areresponsive not only to inXammatory mediators within theCSF space but also to signals released by damaged sensoryaVerents (Wieseler-Frank et al., 2006a). Given both the den-sity of sensory nerve Wbers innervating the meninges(Strassman et al., 2004), whose only known function is torelay pain (Ray and WolV, 1940), plus the intimate contactbetween the meninges and CSF space, these new Wndings


suggest that the meninges may be an under-recognized con-tributor to spinal pain signaling. Consistent with this,inXammation of the arachnoid (arachnoiditis) induces backpain (Petty et al., 2000) as does meningitis (Jeha et al., 2003;Petty et al., 2000).

7. Glial regulation of opioid analgesia, tolerance, and dependence/withdrawal

7.1. Responses of glia to chronic opioids: contribution to morphine tolerance and withdrawal-induced pain enhancement

A strong case has been made that the mechanismsunderlying neuropathic pain and morphine tolerance arestrikingly similar (Mayer et al., 1999). Given the strength ofevidence implicating glia in pain facilitation, including neu-ropathic pain, it was natural to extend this question towhether glia were similarly involved in the production ofmorphine tolerance. As will be reviewed below, it is nowclear that this is indeed the case (Watkins et al., 2005).

The Wrst report linking glia to morphine tolerance dem-onstrated that chronic systemic morphine increased astro-cyte activation in spinal cord (Song and Zhao, 2001).Importantly, co-administration of Xuorocitrate (a glial met-abolic inhibitor) with morphine signiWcantly attenuated notonly glial activation but also morphine tolerance (Song andZhao, 2001). Hence this work associatively and causallylinked the two phenomena.

Work that rapidly followed in our laboratory and othersstrongly supported and extended these initial observations.Chronic morphine was reported to: (a) activate microgliaas well as astrocytes (Cui et al., 2006; Raghavendra et al.,2002; Tai et al., 2006), (b) upregulate TNF, IL-1 and IL-6 inspinal cord (Johnston et al., 2004; Raghavendra et al., 2002;Tai et al., 2006), (c) upregulate TNF, IL-1 and IL-6 inmicroglia but not neurons (Tai et al., 2006), and (d) pro-gressively induce tolerance which was temporally corre-lated with increasing glial activation and proinXammatorycytokine production (Raghavendra et al., 2004). Impor-tantly, morphine tolerance was slowed or reversed by eitherinhibition of spinal proinXammatory cytokines (Johnstonet al., 2004; Raghavendra et al., 2002; Shavit et al., 2005) orby knocking out IL-1 signaling (Shavit et al., 2005). A linkto glial p38 MAPK was reported by Cui et al. (2006) whodemonstrated that: (a) intrathecal administration of mor-phine for 7 days signiWcantly increased the number of phos-pho-p38 MAPK positive cells in spinal cord, (b) thisphospho-p38 MAPK expression was restricted to microg-lia, (c) a speciWc p38 inhibitor attenuated the developmentof tolerance to morphine analgesia. Lastly, a link has alsorecently been made to glial glutamate transporters. Chronicmorphine regimens that induce tolerance have been demon-strated to downregulate spinal cord dorsal horn glialGLAST and GLT-1 glutamate transporters (the majortransporters responsible for regulating extracellular levelsof excitatory amino acids) and concomitantly upregulate

extracellular excitatory amino acids (Mao et al., 2002; Taiet al., 2006). Such data suggest that tolerance may, in part,be created by an opposing increase in neuronal excitabilitydue to increased ongoing stimulation by glutamate. Attenu-ation of morphine tolerance by amitriptyline was associ-ated with upregulation of both GLAST and GLT-1 (andassociated lowering of extracellular excitatory amino acids)as well as blockade of elevated proinXammatory cytokines(Tai et al., 2006). These seemingly independent events infact may not be independent, as proinXammatory cytokinesinhibit glutamate uptake by astrocytes and reduce glialGLAST and GLT-1 expression (Korn et al., 2005; Okadaet al., 2005). Thus, it may be that morphine stimulates glialproinXammatory cytokines that, downstream, downregu-late glial glutamate transporters, thereby enhancing neuro-nal excitability. This would be in addition to the variousneuroexcitatory eVects reviewed above (see Section 5).

The work of Raghavendra et al. (2002, 2004) and ourlaboratory (Johnston et al., 2004) also revealed a role ofglia in pain enhancement that naturally occurs upon cessa-tion of chronically administered opioids; that is, opioidwithdrawal induced pain facilitation. These studiesreported that withdrawal-induced pain enhancement isblocked by: (a) drugs or IL-10 (an anti-inXammatory cyto-kine) that block glial proinXammatory cytokine produc-tion, or (b) IL-1 receptor antagonist (Johnston et al., 2004;Raghavendra et al., 2002; Raghavendra et al., 2004).

Importantly, a link back to neuropathic pain has alsobeen made. What is known from the clinical and laboratoryanimal studies is that morphine loses its eYcacy in neuro-pathic pain patients and rats. That is, a state akin to toler-ance prevails, despite the lack of prior opioid exposure;hence the patients experience naive tolerance rendering opi-oids ineVective. Raghavendra et al. (2002) demonstratedthat glial activation and glial proinXammatory cytokineproduction by morphine was exaggerated in rats with neu-ropathic pain. Indeed, spinal inhibition of proinXammatorycytokines abolished the morphine resistance in neuropathicanimals. That is, in the presence of proinXammatory cyto-kine inhibition, the analgesic eYcacy of even acute mor-phine was restored in rats with neuropathic pain(Raghavendra et al., 2002) and therefore the detrimentaleVects of naive tolerance were ameliorated.

7.2. Responses of glia to acute opioids: opposition of acute opioid analgesia

While the research relating glia to morphine actionbegan with a focus on morphine tolerance, it soon becameapparent that glia modulate the acute analgesic eVects ofmorphine as well (Watkins et al., 2005). We and others haveshown that blockade of spinal proinXammatory cytokinesincreases the magnitude and duration of acute analgesia tomorphine (Johnston et al., 2004; Shavit et al., 2005) andmethadone (Hutchinson et al., 2006a). Similarly, adminis-tration of a neutral dose of IL-1 (i.e., a low dose exertingno detectable eVects on pain responsivity on its own)


abolished, whereas knocking out IL-1 signaling potentiatedand prolonged, morphine analgesia (Shavit et al., 2005).Intriguingly, administration of IL-1 receptor antagonistimmediately upon apparent cessation of morphine analge-sia rapidly “reinstates” analgesia suggesting that morphine-induced glial release of IL-1 “masks” ongoing morphineanalgesia by exerting an opposing eVect (Coats et al., 2006;Hutchinson et al., 2006b; Shavit et al., 2005). Thus, IL-1and potentially other proinXammatory cytokines, opposethe ability of acute morphine to suppress pain.

A parallel conclusion was reached using a very diVerentapproach, namely using sickness-induced pain facilitationto activate spinal cord glia prior to morphine challenge.Johnston and Westbrook (2003, 2005) demonstrated thatspinal glial activation in response to a systemic injection ofLPS on day 1 prevented systemic morphine analgesia onday 2, when no sickness induced pain facilitation remained.This eVect was prevented by spinal administration of theglial inhibitor Xuorocitrate prior to LPS (Johnston andWestbrook, 2005). The reverse case may prove true as well;that is, prior morphine may enhance proinXammatorycytokine production to a later dose of LPS. Data frommicroglial cell cultures clearly indicate that priming gliawith prior 24 hr co-culture with morphine led to signiW-cantly increased release of TNF and superoxide upon LPSchallenge (Chao et al., 1994).

While speculative, the above discussion suggests a solu-tion to what has come to be called: “paradoxical” opioid-induced pain enhancement (King et al., 2005; Mao, 2002).The observation in humans as well as laboratory animals isthat repetitive opioid exposure leads to the development ofabnormal sensitivity to pain. Initial human and laboratoryanimal reports used repeated injections of opioid and henceit was argued that the development of “paradoxical” opi-oid-induced pain enhancement was the result of “mini-withdrawals” occurring between subsequent injections.However, that argument was put to rest in animal studiesemploying osmotic minipumps, for example, so to ensureinvariant dosage of opioid across time. Even here, “para-doxical” opioid-induced pain enhancement began within afew hours of opioid initiation and developed over time(Mao, 2002; Vanderah et al., 2000). As reviewed by Mao,“collectively, these data indicate an important concept thatprolonged opioid treatment not only results in a loss of opi-oid antinociceptive eYcacy but also leads to activation of apronociceptive system manifested as reduction of nocicep-tive thresholds.” To date, evidence points to involvement ofdynorphin release and glutamatergic NMDA receptors(Mao, 2002). Given: (a) the progressive activation of gliawith repeated opioid exposure (including with constantopioid exposure using osmotic minipumps), (b) the evi-dence that dynorphin and NMDA activate glia and inducethe release of proinXammatory cytokines (Laughlin et al.,2000), exploring whether glial activation will solve the “par-adox” of opioid-induced abnormal pain sensitivity wouldseem warranted for both theoretical and practical clinicalreasons.

7.3. Responses of glia to repeated opioids: contribution to opioid dependence/withdrawal

Studies of glial involvement in morphine analgesic toler-ance made one additional intriguing observation. That is,upon cessation of morphine, rats exhibited withdrawal-induced pain enhancement indicating that morphine depen-dence, as well as tolerance, had been established. This wasnot surprising, as dependence develops upon repetitivetreatment with many drugs, including opioids. Dependencerequires continued re-exposure to opioids to avoid a with-drawal syndrome that would otherwise occur when opioiddosing ceases. One sign of opioid withdrawal in rats, as wellas people, is pain enhancement. What was important in thestudies of Raghavendra et al. (2002, 2004) and our labora-tory (Johnston et al., 2004) was that rats that had eitherreceived glial modulatory drugs or intrathecal IL-10 to sup-press proinXammatory cytokine production, or intrathecalIL-1 receptor antagonist throughout the tolerance/depen-dence regimen, showed no such withdrawal induced painenhancement. Thus these data were the Wrst to suggest thatglia may be involved in opioid dependence/withdrawal inaddition to simply tolerance to the pain suppressive actionsof opioids.

To explore glial regulation of dependence/withdrawal,we gave rats repeated systemic administration of morphineto establish morphine dependence. Each morphine dosewas co-administered with AV411, a blood-brain barrierpermeable glial activation inhibitor (Ledeboer et al.,2006b). After the last drug dose, withdrawal was precipi-tated by administering the opioid antagonist naloxone.Rats not receiving AV411 showed robust withdrawal signsover time (e.g., “wet dog” shakes, excessive grooming/penislicking, etc.) whereas AV411-treated rats did not (Lewiset al., 2006). Correlated with this, we found that this sys-temic morphine regimen upregulated astrocytic andmicroglial activation markers throughout the brain andspinal cord, whereas AV411 maintained glial activation atnear basal levels in morphine-treated rats (Lewis et al.,2006). These data clearly suggest that glia, in addition toregulating pain facilitation, opioid analgesia and opioid tol-erance, now must be considered as contributing to the phe-nomenon of morphine dependence/withdrawal as well. Wewould predict that inhibiting glial activation in response toopioids would allow opioids to be more eVectively used forpain management. As fear of opioid dependence/with-drawal is cited as a major reason for underprescribing opi-oids by physicians, under-medicating with opioid by nurses,and avoidance of opioids by patients (Portenoy, 1994), pre-venting glia from driving dependence/withdrawal would bepredicted to have a major positive impact on healthcare.

7.4. Non-classical eVects of opioids in activating glia

A point worth noting here is that it has been assumedthat morphine activates glia via classical opioid receptors,in every journal publication cited above, including our own.


We have recently begun questioning whether this is in facttrue for opioid-induced glial activation. Our strategy hasbeen based on Hutchinson and colleagues’ studies demon-strating that opioids could markedly aVect the function ofperipheral immune cells via non-classical opioid receptors(Hutchinson and Somogyi, 2004a,b, 2005). The key pointhere is that classical opioid receptors, which are the onlyknown opioid receptors on neurons, are stereoselective.That is, [¡]-isomers are active (e.g., they bind the receptorand activate it) whereas [+]-isomers have greatly reducedaYnity and are not. Thus [¡]-methadone produces painsuppression while [+]-methadone is without eVect. What isintriguing in the studies of Hutchinson et al. is that both the[+] and the [¡]-isomers of opioids were equally eVective inaltering the function of peripheral immune cells bothin vitro and in vivo, clearly documenting that a receptorother than the classical opioid receptor must, by deWnition,be responsible for such eVects (Hutchinson and Somogyi,2004a,b, 2005). A structure activity relationship was alsoconducted using over 20 diVerent opioid compounds and itdemonstrated that the eVect on peripheral immune cells didnot occur in a manner that resembled classical opioid char-acteristics.

Extending these observations to glia, we compared theeVects of chronic intrathecal [¡]-morphine and [¡]-metha-done (both active at classical opioid receptors) to that of[+]-methadone (inactive at classical opioid receptors) andsaline. The outcome measure was dorsal spinal cord IL-1gene expression. The experimental outcome was that “inac-tive” [+]-methadone was equally eVective as [¡]-morphineand [¡]-methadone in increasing IL-1 mRNA compared tosaline (Coats et al., 2006). This suggests that diVerent recep-tors mediate the (classical) pain suppressive eVects of mor-phine than its (non-classical) pain enhancing eVects. If thisproves to be true, the clinical implications are enormous.This is because one should then be able to separate thesephenomena and in doing so enhance the ability of mor-phine to suppress pain, reduce the development of mor-phine tolerance, and reduce the development of opioiddependence/withdrawal. This predicts that such clinicallyrelevant outcomes should be attainable either by (a) modi-fying the physical structure of opioids to not obscure theclassical opioid binding site while preventing opioid bind-ing to the (as yet unknown) non-stereoselective receptor or(b) co-administering an [+]-opioid antagonist which blocksglial activation while leaving neuronal activation by opi-oids unaltered. To date we have explored the second optionwith exciting success. We have found that, indeed, [+]-nal-oxone potentiates the magnitude and duration of opioidanalgesia. In addition, upon chronic co-administration withmorphine in a morphine dependence paradigm, [+]-nalox-one delays the development of morphine tolerance, anddecreases [¡]-naloxone-precipitated withdrawal behaviors(Coats et al., 2006). Thus early indications are that this maywell be a clinically relevant strategy for increasing opioidanalgesia while decreasing the negative consequences ofrepeated opioids.

8. How can glial activation aVect neuronal responses to opioids?

This issue is obviously related to the parallel discussionin Section 5, above. As we have shown that proinXamma-tory cytokines are also released in response to morphine(Johnston et al., 2004) the reader is referred back to theprior section for review of issues that are equally pertinenthere. Beyond what has already been discussed, there areadditional mechanisms that may come into play whenexploring how glial activation may suppress responsivity tomorphine.

The Wrst is that opioids can induce the release of chemo-kines, as well as cytokines. For example, morphine inducesthe release of CCL1/MCP-1, CCL5/RANTES and CCL12/SDF-1 (El-Hage et al., 2006). Such chemokines release inspinal cord would be predicted to enhance pain (Abbadie,2005).

An intriguing line of investigation has emerged indicat-ing that chemokines can alter the actions of opioids via amechanism quite distinct from those discussed for proin-Xammatory cytokines, above. This mechanism involvesinactivation of opioid receptor function via heterologousdesensitization. Chemokines, via binding to their cognatereceptors (at least CCR1, CCR5 and CXCR4) on a cell canlead, via intracellular signaling cascades, to the desensitiza-tion/inactivation of opioid receptors expressed on the samecell (Adler et al., 2006; Zhang et al., 2004). Some forms ofchemokine/opioid heterologous desensitization can be bidi-rectional, where opioids can desensitize chemokines recep-tors as well as the reverse. But there appear to be a range ofpossibilities including unidirectional, where only chemo-kines desensitize opioid receptors and not the reverse(Adler et al., 2006). While multiple chemokines can desensi-tize opioid receptors, it is not universal as CCL2/MCP-1has no such actions (Adler et al., 2006).

While Wrst discovered in studies of chemokines/opioidinteractions on peripheral immune cells, the concept hasproven valid for chemokines/opioid interactions on neu-rons as well (Adler et al., 2006). When rats were adminis-tered CXC12/SDF1 or CCL5/RANTES into theperiaqueductal gray (a major site where morphine producespain suppression) 30 min prior to morphine or DAMGO (amu-opioid receptor agonist), analgesia was suppressed(Szabo et al., 2001). As AMD3100, a CXCL12 receptorantagonist, abolished the desensitization, allowing for fullopioid analgesia, this conWrmed that the eVects observedwere mediated through that chemokine receptor (Adleret al., 2006). Such eVects are not restricted to mu opioidreceptors but rather extends equally to delta and kappaopioid receptors as well (Adler et al., 2006).

9. Implications for clinical pain control

The potential implications of glial involvement in thedysregulation of pain and opioid actions are vast. Regardingpain, the fundamental, inescapable fact is that present


treatments for chronic pain fail. The drugs currently used totreat clinical pain problems were all developed to target neu-rons, prior to the discovery that glia regulate pain. Presentknowledge of glial regulation of pain derives from manydiVerent laboratories using many diVerent animal andin vitro models, and equally as many diVerent endpoints.Such a broad consensus cannot be ignored and demandsthat novel therapies be examined in clinical trials to deter-mine whether glially driven pain occurs in humans as it doesin laboratory animals (Watkins and Maier, 2003). While it isdiYcult to change the status quo, when the status quo leavesso many pain patients in misery, it is time to seek change.

Beyond pain facilitation are the new data reviewed here,showing that glia compromise the eYcacy of opioids. Howbroadly this impacts analgesic drugs is currently unknown,but clearly extends beyond morphine as we have observedthe same eVects with methadone as well (Coats et al., 2006).The fact that non-classical opioid receptors appear to medi-ate the glial activating eVects of opioids suggests that theclinical eYcacy of morphine and methadone, at minimum,can be improved in magnitude and duration by separatingthe glial activating eVects of opioids from their pain sup-pressive ones. The data suggest that chemical modiWcationof opioids may be able to retain the portions of the mole-cules that bind to neuronal receptors, while avoiding bind-ing to the (as yet unknown) non-stereoselective receptor(s)that activate glia. Alternatively, one can envision the daywhen a glial activation inhibitor would be co-administeredwith every dose of opioid so as to enhance analgesia, delaythe development of tolerance, and delay the development ofopioid dependence/withdrawal.

Such interventions are needed, not only because clinicalpain syndromes are so poorly controlled by drugs such asopioids, but also because present treatments of opioiddependence/withdrawal fail as well, with extremely highrates of return to drug use (60–80%) (Ball and Ross, 1991;Charles et al., 2003). Dependence is an expected butunwanted consequence of opioid use, and withdrawal fromopioid use is highly aversive both physically and motiva-tionally (Savage, 1999). Dependence can occur rapidly aswithdrawal syndromes can occur in humans after even briefopioid use (»2 days) (Sees and Clark, 1993). This compelshumans and other organisms to seek or use opioid for reliefor avoidance of withdrawal symptoms (Savage, 1999). Theaversive physical and motivational changes that comprisethe withdrawal syndrome are signiWcant factors deterringdrug-dependent humans from going through rehabilitation(Stewart et al., 1984). If preventing glial activation can helpavoid the development of opioid dependence/withdrawal orcould ease withdrawal from ongoing opioid dependencies,then one would predict that opioid use for pain controlwould be far safer for prolonged pain treatments.

References

Abbadie, C., 2005. Chemokines, chemokine receptors and pain. TrendsImmunol. 26, 529–534.

Abbadie, C., Lindia, J.A., Cumiskey, A.M., Peterson, L.B., Mudgett, J.S.,Bayne, E.K., DeMartino, J.A., MacIntyre, D.E., Forrest, M.J., 2003.Impaired neuropathic pain responses in mice lacking the chemokinereceptor CCR2. Proc. Natl. Acad. Sci. USA 100, 7947–7952.

Adler, M.W., Geller, E.B., Chen, X., Rogers, T.J., 2006. Viewing chemo-kines as a third major system of communication in the brain. Aaps J. 7,E865–E870.

Aicher, S.A., Sharma, S., Cheng, P.Y., Pickel, V.M., 1997. The N-methyl-D-aspartate (NMDA) receptor is postsynaptic to substance P-containingaxon terminals in the rat superWcial dorsal horn. Brain Res. 772, 71–81.

Aldskogius, H., Kozlova, E.N., 1998. Central neuron-glial and glial-glialinterations following axon injury. Prog. Neurobiol. 55, 1–26.

Alexander, G.M., van Rijn, M.A., van Hilten, J.J., Perreault, M.J., Sch-wartzman, R.J., 2005. Changes in cerebrospinal Xuid levels of pro-inXammatory cytokines in CRPS. Pain 116, 213–219.

Backonja, M, Muller, D, Coe, C. 2006. Neuropathic pain and levels ofcytokines and IL-10 in the blood and cerebrospinal Xuid. In: Proceed-ings of the11th World Congress on Pain 1647-P150.

Ball, J.C., Ross, A., 1991. Follow-up study of 105 patients who left treat-ment. In: Ball, J.C., Ross, A. (Eds.), The EVectiveness of MethadoneMaintenance Treatment: Patients, Programs, Services & OutcomesSpringer-Verlag, New York.

Besong, G., Battaglia, G., D’Onofrio, M., Di Marco, R., Ngomba, R.T., Storto,M., Castiglione, M., Mangano, K., Busceti, C.L., Nicoletti, F.R., Bacon, K.,Tusche, M., Valenti, O., Conn, P.J., Bruno, V., Nicoletti, F., 2002. Activa-tion of group III metabotropic glutamate receptors inhibits the productionof RANTES in glial cell cultures. J. Neurosci. 22, 5403–5411.

Biber, K., Dijkstra, I., Trebst, C., De Groot, C.J., RansohoV, R.M., Bod-deke, H.W., 2002. Functional expression of CXCR3 in cultured mouseand human astrocytes and microglia. Neuroscience 112, 487–497.

Chao, C.C., Gekker, G., Sheng, W.S., Hu, S., Tsang, M., Peterson, P.K.,1994. Priming eVect of morphine on the production of tumor necrosisfactor-alpha by microglia: implications in respiratory burst activityand human immunodeWciency virus-1 expression. J. Pharmacol. Exp.Ther. 269, 198–203.

Chapman, G., Moores, K., Harrison, D., Campbell, C.A., Stewart, B.R.,Strijbos, P.J.L.M., 2000. Fractalkine cleavage from neuronal mem-branes represents an acute event in the inXammatory response to exci-totoxic brain damage. J. Neurosci. 20 RC87, 1–5.

Charles, B.K., Day, J.E., Rollins, D.E., Andrenyak, D., Ling, W., Wilkins,D.G., 2003. Opiate recidivism in a drug-treatment program: compari-son of hair and urine data. J. Anal. Toxicol. 27, 412–428.

Coats, B.D., Hutchinson, M.R., Watkins, L.R., 2006. Opioid inactive (+)-naloxone potentiates morphine analgesia: further evidence that opioidanalgesia is opposed by opioid induced glial activation. NeuroscienceMeeting Planner. Atlanta, GA: Society for Neuroscience.

Costigan, M., Mannion, R.J., Kendall, G., Lewis, S.E., Campagna, J.A.,Coggeshall, R.E., Meridith-Middleton, J., Tate, S., Woolf, C.J., 1998.Heat shock protein 27: developmental regulation and expression afterperipheral nerve injury. J. Neurosci. 18, 5891–5900.

Cui, Y., Chen, Y., Zhi, J.L., Guo, R.X., Feng, J.Q., Chen, P.X., 2006. Activa-tion of p38 mitogen-activated protein kinase in spinal microglia medi-ates morphine antinociceptive tolerance. Brain Res. 1069, 235–243.

Dame, J.B., Juul, S.E., 2000. The distribution of receptors for the pro-inXammatory cytokines interleukin (IL)-6 and IL-8 in the developinghuman fetus. Early Hum. Dev. 58, 25–39.

Dantzer, R., 2004. Cytokine-induced sickness behaviour: a neuroim-mune response to activation of innate immunity. Eur. J. Pharmacol.500, 399–411.

De, A., Krueger, J.M., Simasko, S.M., 2003. Tumor necrosis factor alphaincreases cytosolic calcium responses to AMPA and KCl in primarycultures of rat hippocampal neurons. Brain Res. 981, 133–142.

El-Hage, N., Wu, G., Ambati, J., Bruce-Keller, A.J., Knapp, P.E., Hauser,K.F., 2006. CCR2 mediates increases in glial activation caused byexposure to HIV-1 Tat and opiates. J. Neuroimmunol.

Falchi, M., Ferrara, F., Gharib, C., Dib, B., 2001. Hyperalgesic eVect ofintrathecally administered interleukin-1 in rats. Drugs Exp. Clin. Res.27, 97–101.


Grassi, F., Mileo, A.M., Monaco, L., Punturieri, A., Santoni, A., Eusebi, F.,1994. TNF-alpha increases the frequency of spontaneous miniaturesynaptic currents in cultured rat hippocampal neurons. Brain Res. 659,226–230.

Hains, B.C., Waxman, S.G., 2006. Activated microglia contribute to themaintenance of chronic pain after spinal cord injury. J. Neurosci. 26,4308–4317.

Harris, H.E., Andersson, U., 2004. The nuclear protein HMGB1 as a pro-inXammatory mediator. Eur. J. Immunol. 34, 1503–1512.

Hashizume, H., Rutkowski, M.D., Weinstein, J.N., DeLeo, J.A., 2000. Cen-tral administration of methotrexate reduces mechanical allodynia in ananimal model of radiculopathy/sciatica. Pain 87, 159–169.

Haydon, P.G., Carmignoto, G., 2006. Astrocyte control of synaptic trans-mission and neurovascular coupling. Physiol. Rev. 86, 1009–1031.

Holguin, A., Biedenkapp, J., Campisi, J., Wieseler-Frank, J., O’Connor,K.A., Milligan, E.D., Maksimova, E., Bravmann, C., Hansen, M.K.,Martin, D., Fleshner, M., Maier, S.F., Watkins, L.R., 2004. HIV-1gp120 stimulates proinXammatory cytokine-mediated pathologicalpain via activation of nitric oxide synthase-I (nNOS). Pain 110, 517–530.

Holmes, G.M., Hebert, S.L., Rogers, R.C., Hermann, G.E., 2004. Immuno-cytochemical localization of TNF type 1 and type 2 receptors in the ratspinal cord. Brain Res. 1025, 210–219.

Honore, P., Wade, C.L., Zhong, C., Harris, R.R., Wu, C., Ghayur, T., Iwak-ura, Y., Decker, M.W., Faltynek, C., Sullivan, J., Jarvis, M.F., 2006.Interleukin-1alphabeta gene-deWcient mice show reduced nociceptivesensitivity in models of inXammatory and neuropathic pain but notpost-operative pain. Behav. Brain Res. 167, 355–364.

Hutchinson, M.R., Somogyi, A.A., 2004a. Relationship between 4,5-epox-ymorphinan structure and in vitro modulation of cell proliferation.Eur. J. Pharmacol. 494, 251–262.

Hutchinson, M.R., Somogyi, A.A., 2004b. (S)-(+)-methadone is moreimmunosuppressive than the potent analgesic (R)-(–)-methadone. Int.Immunopharmacol. 4, 1525–1530.

Hutchinson, M.R., Somogyi, A.A., 2005. Characterisation of the in vitromodulation of splenocyte proliferation by non-4,5-epoxymorphinanopioids. Int. Immunopharmacol. 5, 1713–1722.

Hutchinson, MR, Coats, BD, Lewis, SS, Maier, SF, Watkins, LR. SpinalproinXammatory cytokines oppose acute opioid analgesia followingintrathecal and subcutaneous opioid administration. Proc Soc Neuro-sci, in press.

Hutchinson, M.R., Milligan, E.D., Jekich, B.M., Coats, B.D., Lewis, S.,Maier, S.F., Watkins, L.R., 2006b. Elevations of spinal proinXamma-tory cytokines are mediated via non-classical opioid mechanisms. Proc.Soc. NeuroImmune Pharmacol..

Hwang, D., 2001. Modulation of the expression of cyclooxygenase-2 byfatty acids mediated through toll-like receptor 4-derived signalingpathways. Faseb J. 15, 2556–2564.

Inoue, K., 2006. The function of microglia through purinergic receptors:neuropathic pain and cytokine release. Pharmacol. Ther. 109, 210–226.

Inoue, M., Rashid, M.H., Fujita, R., Contos, J.J., Chun, J., Ueda, H., 2004.Initiation of neuropathic pain requires lysophosphatidic acid receptorsignaling. Nat. Med. 10, 712–718.

Jeha, L.E., Sila, C.A., Lederman, R.J., Prayson, R.A., Isada, C.M., Gordon,S.M., 2003. West Nile virus infection: a new acute paralytic illness.Neurology 61, 55–59.

John, G.R., Lee, S.C., Song, X., Rivieccio, M., Brosnan, C.F., 2005. IL-1-regulated responses in astrocytes: relevance to injury and recovery.Glia 49, 161–176.

Johnston, I.N., Westbrook, R.F., 2003. Acute and conditioned sicknessreduces morphine analgesia. Behav. Brain Res. 142, 89–97.

Johnston, I.N., Westbrook, R.F., 2005. Inhibition of morphine analgesiaby LPS: role of opioid and NMDA receptors and spinal glia. Behav.Brain Res. 156, 75–83.

Johnston, I.N., Milligan, E.D., Wieseler-Frank, J., Frank, M.G., Zapata, V.,Campisi, J., Langer, S., Martin, D., Green, P., Fleshner, M., Leinwand,L., Maier, S.F., Watkins, L.R., 2004. A role for proinXammatory cyto-kines and fractalkine in analgesia, tolerance, and subsequent pain facil-

itation induced by chronic intrathecal morphine. J. Neurosci. 24, 7353–7365.

King, T., Ossipov, M.H., Vanderah, T.W., Porreca, F., Lai, J., 2005. Is para-doxical pain induced by sustained opioid exposure an underlyingmechanism of opioid antinociceptive tolerance? Neurosignals 14, 194–205.

Kleibeuker, W, Ledeboer, A, Watkins, LR, Maier, SF, Zijlstra, J, Heijnen,CJ, Kavelaars, A. 2006. Neuronal G protein-coupled receptor kinase 2is downregulated in the rat spinal during chronic constriction injury. JNeurochem, in review.

Korn, T., Magnus, T., Jung, S., 2005. Autoantigen speciWc T cells inhibitglutamate uptake in astrocytes by decreasing expression of astrocyticglutamate transporter GLAST: a mechanism mediated by tumornecrosis factor-alpha. Faseb J. 19, 1878–1880.

Kreutzberg, G.W., 1996. Microglia: a sensor for pathological events in theCNS. Trends Neurosci. 19, 312–318.

Laughlin, T.M., Bethea, J.R., Yezierski, R.P., Wilcox, G.L., 2000. Cytokineinvolvement in dynorphin-induced allodynia. Pain 84, 159–167.

Ledeboer, A., Sloane, E.M., Milligan, E.D., Frank, M.G., Mahony, J.H.,Maier, S.F., Watkins, L.R., 2005. Minocycline attenuates mechanicalallodynia and proinXammatory cytokine expression in rat models ofpain facilitation. Pain 115, 71–83.

Ledeboer, A, Liu, T, Shumilla, JA, Mahoney, JH, Vargas, J.A., Sultzbaugh,L., Sanftner, L.M., Watkins, L.R.Johnson, K.W., 2006a. The glial mod-ulatory drug AV411 attenuates mechanical allodynia in rat models ofneuropathic pain. Pain., in review.

Ledeboer, A., Mahoney, J.H., Milligan, E.D., Martin, D., Maier, S.F., Wat-kins, L.R., 2006b. Spinal cord glia and interleukin-1 do not appear tomediate persistent allodynia induced by intramuscular acidic saline inrats. J. Pain 7, 757–767.

Lee, J.C., Mayer-Proschel, M., Rao, M.S., 2000. Gliogenesis in the centralnervous system. Glia 30, 105–121.

Lehnardt, S., Lachance, C., Patrizi, S., Lefebvre, S., Follett, P.L., Jensen,F.E., Rosenberg, P.A., Volpe, J.J., Vartanian, T., 2002. The toll-likereceptor TLR4 is necessary for lipopolysaccharide-induced oligoden-drocyte injury in the CNS. J. Neurosci. 22, 2478–2486.

Lewis, S.S., Maier, S.F., Watkins, L.R., 2006. Spinal pro-inXammatorycytokines oppose acute opioid analgesia following intrathecal & subcu-taneous opioid administration. Neuroscience Meeting Planner.Atlanta, GA: Society for Neuroscience.

Lewis, S.S., Hutchinson, M.R., Coats, B.D., Brzeski, A.L., Maier, S.F., Wat-kins, L.R., Johnson, K.W., 2006. AV411, a blood brain barrier perme-able glial activation inhibitor, reduces morphine withdrawal behaviorsin rats. Neuroscience Meeting Planner. Atlanta,GA: Society for Neuro-science.

Lindia, J.A., McGowan, E., Jochnowitz, N., Abbadie, C., 2005. Inductionof CX3CL1 expression in astrocytes and CX3CR1 in microglia in thespinal cord of a rat model of neuropathic pain. J. Pain 6, 434–438.

Loredo, G.A., Benton, H.P., 1998. ATP and UTP activate calcium-mobliz-ing P2U-like receptors and act synergistically with interleukin-1 tostimulate prostaglandin E2 release from human rheumatoid synovialcells. Arthritis Rheum. 41, 246–255.

Maier, S.F., Watkins, L.R., 1998. Cytokines for psychologists: implicationsof bidirectional immune-to-brain communication for understandingbehavior, mood, and cognition. Psychol. Rev. 105, 83–107.

Mao, J., 2002. Opioid-induced abnormal pain sensitivity: implications inclinical opioid therapy. Pain 100, 213–217.

Mao, J., Sung, B., Ji, R.R., Lim, G., 2002. Chronic morphine induces down-regulation of spinal glutamate transporters: implications in morphinetolerance and abnormal pain sensitivity. J. Neurosci. 22, 8312–8323.

Marriott, D.R., Wilkin, G.P., Wood, J.N., 1991. Substance P-induced releaseof prostaglandins from astrocytes: regional specialisation and correlationwith phosphoinositol metabolism. J. Neurochem. 56, 259–265.

Mayer, D.J., Mao, J., Holt, J., Price, D.D., 1999. Cellular mechanisms ofneuropathic pain, morphine tolerance, and their interactions. Proc.Natl. Acad. Sci. USA 96, 7731–7736.

McMahon, S.B., CaVerty, W.B., Marchand, F., 2005. Immune and glial cellfactors as pain mediators and modulators. Exp. Neurol. 192, 444–462.


Meller, S.T., Dyskstra, C., Grzybycki, D., Murphy, S., Gebhart, G.F., 1994.The possible role of glia in nociceptive processing and hyperalgesia inthe spinal cord of the rat. Neuropharmacology 33, 1471–1478.

Milligan, E.D., Zapata, V., Chacur, M., Schoeniger, D., Biedenkapp, J.,O’Connor, K.A., Verge, G.M., Chapman, G., Green, P., Foster, A.C.,Naeve, G.S., Maier, S.F., Watkins, L.R., 2004. Evidence that exogenousand endogenous fractalkine can induce spinal nociceptive facilitationin rats. Eur. J. Neurosci. 20, 2294–2302.

Milligan, E., Zapata, V., Schoeniger, D., Chacur, M., Green, P., Poole, S.,Martin, D., Maier, S.F., Watkins, L.R., 2005. An initial investigation ofspinal mechanisms underlying pain enhancement induced by fractal-kine, a neuronally released chemokine. Eur. J. Neurosci. 22, 2775–2782.

Milligan, E.D., Sloane, E.M., Langer, S.J., Hughes, T.S., Jekich, B.M.,Frank, M.G., Mahoney, J.H., LevkoV, L.H., Maier, S.F., Cruz, P.E.,Flotte, T.R., Johnson, K.W., Mahoney, M.M., Chavez, R.A., Leinwand,L.A., Watkins, L.R., 2006. Repeated intrathecal injections of plasmidDNA encoding interleukin-10 produce prolonged reversal of neuro-pathic pain. Pain [Epub ahead of print]. PMID:16949747.

Morioka, N., Inoue, A., Hanada, T., Kumagai, K., Takeda, K., Ikoma, K.,Hide, I., Tamura, Y., Shiomi, H., Dohi, T., Nakata, Y., 2002. Nitricoxide synergistically potentiates interleukin-1 beta-induced increase ofcyclooxygenase-2 mRNA levels, resulting in the facilitation of sub-stance P release from primary aVerent neurons: involvement of cGMP-independent mechanisms. Neuropharmacology 43, 868–876.

Obata, H., Eisenach, J.C., Hussain, H., Bynum, T., Vincler, M., 2006. Spinalglial activation contributes to postoperative mechanical hypersensitiv-ity in the rat. J. Pain. 7, 816–822.

O’Connor, K.A., Hansen, M.K., Rachal Pugh, C., Deak, M.M., Bieden-kapp, J.C., Milligan, E.D., Johnson, J.D., Wang, H., Maier, S.F., Tracey,K.J., Watkins, L.R., 2003. Further characterization of high mobilitygroup box 1 (HMGB1) as a proinXammatory cytokine: central ner-vous system eVects. Cytokine 24, 254–265.

Ohtori, S., Takahashi, K., Moriya, H., Myers, R.R., 2004. TNF-alpha andTNF-alpha receptor type 1 upregulation in glia and neurons afterperipheral nerve injury: studies in murine DRG and spinal cord. Spine29, 1082–1088.

Okada, K., Yamashita, U., Tsujis, S., 2005. Modulation of Na+-dependentglutamate transporter of murine astocytes by inXammatory mediators.J. UOEH 27, 161–170.

Oka, T., Aou, S., Hori, T., 1994. Intracerebroventricular injection of inter-leukin-1b enhances nociceptive neuronal responses of the trigeminalnucleus caudalis in rats. Brain Res. 656, 236–244.

Owens, T., Babcock, A.A., Millward, J.M., Toft-Hansen, H., 2005. Cyto-kine and chemokine inter-regulation in the inXamed or injured CNS.Brain Res. Brain Res. Rev. 48, 178–184.

Owolabi, S.A., Saab, C.Y., 2006. Fractalkine and minocycline alter neuro-nal acivity in the spinal cord dorsal horn. FEBS Lett. 580, 4306–4310.[Epub ahead of print].

Palma, C., Minghetti, L., AtolW, M., Ambrosini, E., Silberstein, F.C., manz-ini, S., Levi, G., Aloisi, F., 1997. Functional characterization of sub-stance P receptors on cultured human spinal cord astrocytes:synergism of substance P with cytokines in inducing interleukin-6 andprostaglandin E2 production. Glia 21, 183–193.

Park, J.S., Gamboni-Robertson, F., He, Q., Svetkauskaite, D., Kim, J.Y.,Strassheim, D., Sohn, J.W., Yamada, S., Maruyama, I., Banerjee, A.,Ishizaka, A., Abraham, E., 2006. High mobility group box 1 proteininteracts with multiple Toll-like receptors. Am. J. Physiol. Cell Physiol.290, C917–C924.

Pasti, L., Zonata, M., Pozzan, T., Vicini, S., Carmignoto, G., 2001. Cyto-solic calcium oscillations in astrocytes may regulate exocytotic releaseof glutamate. J. Neurosci. 21, 477–484.

Perea, G., Araque, A., 2002. Communication between astrocytes and neu-rons: a complex language. J. Physiol. 96, 199–207.

Perry, V.H., Hume, D.A., Gordon, S., 1985. Immunohistochemical localisa-tion of macrophages and microglia in the adult and developing mousebrain. Neurosci. 15, 313–326.

Petty, P.G., Hudgson, P., Hare, W.S., 2000. Symptomatic lumbar spinalarachnoiditis: fact or fallacy? J. Clin. Neurosci. 7, 395–399.

Portenoy, R.K., 1994. Tolerance to opioid analgesics: clinical aspects. Can-cer Surv. 21, 49–65.

Raghavendra, V., Rutkowski, M.D., DeLeo, J.A., 2002. The role of spinalneuroimmune activation in morphine tolerance/hyperalgesia in neuro-pathic and sham-operated rats. J. Neurosci. 22, 9980–9989.

Raghavendra, V., Tanga, F., DeLeo, J.A., 2003. Inhibition of microglialactivation attenuates the development but not existing hypersensitivityin a rat model of neuropathy. J. Pharmacol. Exp. Ther. 306, 624–630.

Raghavendra, V., Tanga, F.Y., DeLeo, J.A., 2004. Attenuation of mor-phine tolerance, withdrawal-induced hyperalgesia, and associated spi-nal inXammatory immune responses by propentofylline in rats.Neuropsychopharmacology 29, 327–334.

Raivich, G., 2005. Like cops on the beat: the active role of resting microg-lia. Trends Neurosci. 28, 571–573.

Rappert, A., Biber, K., Nolte, C., Lipp, M., Schubel, A., Lu, B., Gerard,N.P., Gerard, C., Boddeke, H.W., Kettenmann, H., 2002. Secondarylymphoid tissue chemokine (CCL21) activates CXCR3 to trigger a Cl-current and chemotaxis in murine microglia. J. Immunol. 168, 3221–3226.

Rappert, A., Bechmann, I., Pivneva, T., Mahlo, J., Biber, K., Nolte, C.,Kovac, A.D., Gerard, C., Boddeke, H.W., Nitsch, R., Kettenmann, H.,2004. CXCR3-dependent microglial recruitment is essential for den-drite loss after brain lesion. J. Neurosci. 24, 8500–85009.

Ray, B.S., WolV, H.G., 1940. Experimental studies on headache: pain-sen-sitive structures of the head and their signiWcance in headache. Arch.Surg. 41, 813–856.

Reeve, A.J., Patel, S., Fox, A., Walker, K., Urban, L., 2000. Intrathecallyadministered endotoxin or cytokines produce allodynia, hyperalgesiaand changes in spinal cord neuronal responses to nociceptive stimuli inthe rat. Eur. J. Pain 4, 247–257.

Samad, T.A., Wang, H., Broom, D.C., Woolf, C.J., 2004. Central neuroim-mune interactions after peripheral inXammation: interleukin-1b poten-tiates synaptic transmission in the spinal cord. Proc. Soc. Neurosci. 34,511.7.

Savage, S.R., 1999. Opioid therapy of chronic pain: assessment of conse-quences. Acta Anaesthesiol. Scand. 43, 909–917.

Scholz, J., Broom, D.C., Youn, D.H., Mills, C.D., Kohno, T., Suter, M.R.,Moore, K.A., Decosterd, I., Coggeshall, R.E., Woolf, C.J., 2005. Block-ing caspase activity prevents transsynaptic neuronal apoptosis and theloss of inhibition in lamina II of the dorsal horn after peripheral nerveinjury. J. Neurosci. 25, 7317–7323.

Sees, K.L., Clark, H.W., 1993. Opioid use in the treatment of chronic pain:assessment of addiction. J. Pain Symp. Manage. 8, 257–264.

Shan, S., Hong, C., Mei, H., Ting-Ting, L., Hai-Li, P., Yu-Qiu, Z. 2006.New evidence for the involvement of spinal fractalkine receptor in painfacilitation and spinal glia.

Shavit, Y., Wolf, G., Goshen, I., Livshits, D., Yirmiya, R., 2005. Interleu-kin-1 antagonizes morphine analgesia and underlies morphine toler-ance. Pain 115, 50–59.

Shubayev, V.I., Myers, R.R., 2002. Anterograde TNF alpha transport fromrat dorsal root ganglion to spinal cord and injured sciatic nerve. Neu-rosci. Lett. 320, 99–101.

Song, P., Zhao, Z.Q., 2001. The involvement of glial cells in the develop-ment of morphine tolerance. Neurosci. Res. 39, 281–286.

Spataro, L.E., Young, J., Wieseler-Frank, J., Milligan, E.D., Jekich, B.M.,Springer, D.B., Maier, S., Fleshner, M., Watkins, L.R., 2006. Abdomi-nary surgery primes interleukin-1 and pain responses to low-dose lipo-polysaccharide (LPS) and cyclophospamide (CP). Proc. Soc. Neurosci.,in press.

Sperlagh, B., Baranyi, M., Hasko, G., Vizi, E.S., 2004. Potent eVect of inter-leukin-1 beta to evoke ATP and adenosine release from rat hippocam-pal slices. J. Neuroimmunol. 15, 33–39.

Stellwagen, D., Beattie, E.C., Seo, J.Y., Malenka, R.C., 2005. DiVerentialregulation of AMPA receptor and GABA receptor traYcking bytumor necrosis factor-alpha. J. Neurosci. 25, 3219–3228.

Stewart, J., de Wit, H., Eikelboom, R., 1984. Role of unconditioned andconditioned drug eVects in the self-administration of opiates and stim-ulants. Psychol. Rev. 91, 251–268.


Stievano, L., Piovan, E., Amadori, A., 2004. C and CX3C chemokines: cellsources and physiopathological implications. Crit. Rev. Immunol. 24,205–228.

Strassman, A.M., Weissner, W., Williams, M., Ali, S., Levy, D., 2004. Axondiameters and intradural trajectories of the dural innervation in the rat.J. Comp. Neurol. 473, 364–376.

Sunnemark, D., Eltayeb, S., Nilsson, M., Wallstrom, E., Lassmann, H., Ols-son, T., Berg, A.L., Ericsson-Dahlstrand, A., 2005. CX3CL1 (fractal-kine) and CX3CR1 expression in myelin oligodendrocyteglycoprotein-induced experimental autoimmune encephalomyelitis:kinetics and cellular origin. J. NeuroinXammation 2, 17.

Svensson, C.I., Hua, X.Y., Powell, H.C., Lai, J., Porreca, F., Yaksh, T.L.,2005. Prostaglandin E2 release evoked by intrathecal dynorphin isdependent on spinal p38 mitogen activated protein kinase. Neuropep-tides 39, 485–494.

Szabo, I., Chen, X.-H., Xin, L., Adler, M.W., Howard, O.M.Z., Oppenheim,J.J., Rogers, T.J., 2001. Heterologous desensitization of opioid recep-tors by chemokines inhibits chemotaxis and enhances the perception ofpain. Proc. Natl. Acad. Sci. 99, 01276–10281.

Tadano, T., Namioka, M., Nakagawasai, O., Tan-No, K., Matsushima, K.,Endo, Y., Kisara, K., 1999. Induction of nociceptive responses by intra-thecal injection of interleukin-1 in mice. Life Sci. 65, 255–261.

Tai, Y.H., Wang, Y.H., Wang, J.J., Tao, P.L., Tung, C.S., Wong, C.S., 2006.Amitriptyline suppresses neuroinXammation and up-regulates gluta-mate transporters in morphine-tolerant rats. Pain.

Takeda, K., Sawamura, S., Tamai, H., Sekiyama, H., Hanaoka, K., 2005.Role for cyclooxygenase 2 in the development and maintenance of neu-ropathic pain and spinal glial activation. Anesthesiology 103, 837–844.

Tanaka, T., Minami, M., Nakagawa, T., Satoh, M., 2004. Enhanced pro-duction of monocyte chemoattractant protein-1 in the dorsal root gan-glia in a rat model of neuropathic pain: possible involvement in thedevelopment of neuropathic pain. Neurosci. Res. 48, 463–469.

Tanga, F.Y., Raghavendra, V., DeLeo, J.A., 2004. Quantitative real-time RT-PCR assessment of spinal microglial and astrocytic activation markers ina rat model of neuropathic pain. Neurochem. Int. 45, 397–407.

Tanga, F.Y., Nutile-McMenemy, N., DeLeo, J.A., 2005. The CNS role ofToll-like receptor 4 in innate neuroimmunity and painful neuropathy.Proc. Natl. Acad. Sci. USA 102, 5856–5861.

TawWk, V.L., Lacroix-Fralish, M.L., Bercury, K.K., Nutile-McMenemy,N., Harris, B.T., Deleo, J.A., 2006. Induction of astrocyte diVerentiationby propentofylline increases glutamate transporter expression in vitro:Heterogeneity of the quiescent phenotype. Glia.

TawWk, V.L., Nutile-McMenemy, N., LaCroix-Fralish, M.L., DeLeo, J.A.,2006. EYcacy of propentofylline, a glial modulating agent, on existingmechanical allodynia following peripheral nerve injury. Brain Behav.Immun., in press, doi:10.1016/j.bbi.2006.07.001.

Tikka, T.M., Koistinaho, J.E., 2001. Minocycline provides neuroprotectionagainst N-methyl-D-aspartate neurotoxicity by inhibiting microglia. J.Immunol. 166, 7527–7533.

Tsacopoulos, M., 2002. Metabolic signaling between neurons and glialcells: a short review. J. Physiol. 96, 283–288.

Tsuda, M., Shigemoto-Mogami, Y., Koizumi, S., Mizokoshi, A., Kohsaka,S., Salter, M.W., Inoue, K., 2003. P2X4 receptors induced in spinalmicroglia gate tactile allodynia after nerve injury. Nature 424, 778–783.

Vanderah, T.W., Gardell, L.R., Burgess, S.E., Ibrahim, M., Dogrul, A.,Zhong, C.M., Zhang, E.T., Malan Jr., T.P., Ossipov, M.H., Lai, J., Por-reca, F., 2000. Dynorphin promotes abnormal pain and spinal opioidantinociceptive tolerance. J. Neurosci. 20, 7074–7079.

Vargas, M.E., Singh, S.J., Barres, B.A. 2006. Why is Wallerian degenerationso slow in the CNS? Glia in Health and Disease, Cold Spring HarborLaboratories, Cold Spring Harbor Laboratories.

Verge, G.M., Milligan, E.D., Maier, S.F., Watkins, L.R., Naeve, G.S., Fos-ter, A.C., 2004. Fractalkine (CX3CL1) and fractalkine receptor(CX3CR1) distribution in spinal cord and dorsal root ganglia underbasal and neuropathic pain conditions. Eur. J. Neurosci. 20, 1150–1160.

Viviani, B., Bartesaghi, S., Gardoni, F., Vezzani, A., Behrens, M.M., Bart-fai, T., Binaglia, M., Corsini, E., Di Luca, M., Galli, C.L., Marinovich,M., 2003. Interleukin-1beta enhances NMDA receptor-mediated intra-cellular calcium increase through activation of the Src family ofkinases. J. Neurosci. 23, 8692–8700.

Watkins, L.R., Maier, S.F., 2000. The pain of being sick: implications ofimmune-to-brain communication for understanding pain. Annu. Rev.Psychol. 51, 29–57.

Watkins, L.R., Maier, S.F., 2003. Glia: a novel drug discovery target forclinical pain. Nat. Rev. Drug Discov. 2, 973–985.

Watkins, L.R., Martin, D., Ulrich, P., Tracey, K.J., Maier, S.F., 1997. Evi-dence for the involvement of spinal cord glia in subcutaneous formalininduced hyperalgesia in the rat. Pain 71, 225–235.

Watkins, L.R., Hansen, M.K., Nguyen, K.T., Lee, J.E., Maier, S.F., 1999.Dynamic regulation of the proinXammatory cytokine, interleukin-1beta: molecular biology for non-molecular biologists. Life Sci. 65, 449–481.

Watkins, L.R., Hutchinson, M.R., Johnston, I.N., Maier, S.F., 2005. Glia:novel counter-regulators of opioid analgesia. Trends Neurosci. 28,661–669.

Watkins, L.R., Wieseler-Frank, J., Milligan, E.D., Johnston, I., Maier, S.F.,in press. Contribution of glia to pain processing in health and disease.Handbook of Clinical Neurology. F. Cervero. New York.

White, F.A., Sun, J., Waters, S.M., Ma, C., Ren, D., Ripsch, M., SteXik, J.,Cortright, D.N., Lamotte, R.H., Miller, R.J., 2005. Excitatory monocytechemoattractant protein-1 signaling is up-regulated in sensory neuronsafter chronic compression of the dorsal root ganglion. Proc. Natl.Acad. Sci. USA 102, 14092–14097.

Wieseler-Frank, J., Kwiecien, O., Jekich, B., Maier, S.F., Watkins, L.R.,2004. Putative neuron-to-glia signals synergize to enhance interleukin-1 production by rat dorsal spinal cord glial cells in vitro. Proc. Soc.Neurosci. 34.

Wieseler-Frank, J., Jekich, B.M., Bland, S., Maier, S.F., Watkins, L.R.,2006a. A novel immune-to CNS communication pathway: cells of themeninges surrounding the spinal cord CSF space produce proinXam-matory cytokines in response to an inXammatory stimulus. BrainBehav. Immun., in press, doi:10.1016/j.bbi.2006.07.004.

Wieseler-Frank, J., Jekich, B.M., Mahoney, J.H., Lewis, S., Coats, B.D.,Maier, S.F., Watkins, L.R., 2006b. Spinal cord meninges: a player inexaggerated pain states? Neuroscience Meeting Planner. Atlanta, GA:Society for Neuroscience.

Wolf, G., Yirmiya, R., Goshen, I., Iverfeldt, K., Holmlund, L., Takeda, K.,Shavit, Y., 2003. Impairment of interleukin-1 (IL-1) signaling reducesbasal pain sensitivity in mice: genetic, pharmacological and develop-mental aspects. Pain 104, 471–480.

Wolf, G., Gabay, E., Tal, M., Yirmiya, R., Shavit, Y., 2006. Genetic impair-ment of interleukin-1 signaling attenuates neuropathic pain, autotomy,and spontaneous ectopic neuronal activity, following nerve injury inmice. Pain 120, 315–324.

Zhang, J., De Koninck, Y., 2006. Spatial and temporal relationshipbetween monocyte chemoattractant protein-1 expression and spinalglial activation following peripheral nerve injury. J. Neurochem. 97,772–783.

Zhang, N., Rogers, T.J., Caterina, M., Oppenheim, J.J., 2004. ProinXam-matory chemokines, such as C–C chemokine ligand 3, desensitizemu-opioid receptors on dorsal root ganglia neurons. J. Immunol. 173,594–599.

http://dx.doi.org/10.1016/j.bbi.2006.07.001




g lia as th e Òb ad gu ysÓ: im p licatio n s fo r im p ro vin g clin ical …ibg · 2009. 1....

Documents