reviews - semantic scholar...integrated central nervous system (cns) process-ing. the cns comprises...

The Role of Endocannabinoid Signalingin Motor Control

Cannabinoid receptors and endocannabinoid signaling are distributed

throughout the rostrocaudal neuraxis. Retrograde signaling via endocannabi-

noid mediates synaptic plasticity in many regions in the central nervous

system. Here, we review the role of endocannabinoid signaling in different

parts of the vertebrate motor system from networks responsible for the

execution of movement to planning centers in the basal ganglia and cortex.

The ubiquity of endocannabinoid-mediated plasticity suggests that it plays an

important role in producing motion from defined circuitries and also for

reconfiguring networks to learn new motor skills. The long-term plasticity

induced by endocannabinoids may provide a long-term buffer that stabilizes

the organization of motor circuits and their activity.

A. El Manira and A. KyriakatosDepartment of Neuroscience, Karolinska Institutet,

Stockholm, [email protected]

Motor behavior represents the overt expression ofintegrated central nervous system (CNS) process-ing. The CNS comprises a number of networks,each controlling a defined motor function. Thefinal motor output is the result of a sophisticatedintegration of their activity. These networks displaya continuous plasticity to allow their activity toadapt to changes in the internal and external en-vironments (7). These mechanisms allow animalsto acquire new motor skills and to reorganize theremaining circuitry to recover function after injuryor illness. One major substrate of the plasticity ofneural circuits is the synaptic communication be-tween the constituent neurons that can bechanged by modulatory systems.

Neuromodulation of synaptic connections hasbeen classically considered to depend on release ofmodulatory transmitters from presynaptic termi-nals that can act postsynaptically on axon termi-nals or dendrites to change synaptic strength (10,11, 22, 35, 53, 71). This view has been revisedfollowing the discovery of endocannabinoid andnitric oxide signaling. Endocannabinoids are lipidmolecules that can be released in a nonsynapticfashion from postsynaptic neurons to travel backonto presynaptic terminals and act as retrogrademessengers (27, 55, 69). They represent a promi-nent example of retrograde signaling in synapticplasticity in many regions of the CNS. Most studiesof endocannabinoid-mediated plasticity in thebrain have concentrated on mechanisms related tomemory and learning (2, 3, 13, 17, 20, 42). Thesestudies have, however, been confronted with thedifficulty of being able to quantify cognitivechanges to link synaptic plasticity to behavior.

A major advantage of motor circuits is that theiroutputs can be readily measured and correlated tothe motor behavior. This places them in a positionto link the synaptic plasticity mediated by endo-cannabinoids to changes that occur in the opera-tion of the network as a whole. An example is thespinal network that generates the basic locomotorpattern and also acts as a processing interface toadjust its output in response to influences from thebrain and sensory afferents (15, 16, 26). Becauseendocannabinoids can in principle be releasedfrom all neurons in the spinal locomotor networks,they provide an important modulatory mechanismthat is not only involved in fine-tuning of the on-going activity but that also may be necessary forthe generation of the locomotor activity.

In this review, we will first provide backgroundon endocannabinoid signaling system by describ-ing their synthesis and degradation, and the recep-tor types they activate. We will then give anaccount of the role of endocannabinoid-mediatedsynaptic plasticity in vertebrate motor systemsproceeding from the spinal cord, where the finalprocessing and execution of movement takesplace, to the planning centers in the basal gangliaand cortex.

Endocannabinoids

Endocannabinoids are lipid molecules principallyderived from membrane phospholipids. Unlikeclassical neurotransmitters and neuropeptides, en-docannabinoids are not stored in vesicles in axonterminals, but rather they are synthesized on de-mand in somata and dendrites. They are subse-quently released from cells and then exert an

REVIEWSPHYSIOLOGY 25: 230–238, 2010; doi:10.1152/physiol.00007.2010

1548-9213/10 ©2010 Int. Union Physiol. Sci./Am. Physiol. Soc.230

by 10.220.33.2 on October 27, 2017

http://physiologyonline.physiology.org/D

ownloaded from

http://physiologyonline.physiology.org/

immediate action as signaling molecules (8, 56).Two major endocannabinoids have been identifiedin the CNS: arachidonoylethanolamide (AEA),commonly known as anandamide, and 2-arachido-noylglycerol (2-AG) that are synthesized and de-graded by separate pathways (FIGURE 1).

The major route for the biosynthesis of anand-amide is via the precursor N-arachidonoyl phos-phatidylethanolamine (NAPE), which is generatedby the enzyme N-acyltransferase in a calcium-de-pendent manner. Anandamide is then generatedby hydrolysis of NAPE by a phospholipase D(NAPE-PLD) (8, 56, 63, 68). Thus the endocannabi-noid anandamide seems to be produced “on de-mand” and released in an activity-dependentmanner by enzymatic cleavage of lipid precursors.The biological inactivation of anandamide ismainly through hydrolyzation mediated by fattyacid amide hydrolase (FAAH) (47). The major path-way for the biosynthesis of 2-AG comprises se-quential hydrolysis of arachidonic acid-containinginositol phospholipids by phospholipase C (PLC)and diacylglycerol lipase (8, 56, 63). In response tomany extracellular signals such as neurotransmitters,PLC catalyzes the hydrolysis of phosphatidylinositol-4,5-bisphosphate (PIP2), thereby generating two wellestablished second messengers, inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). DAGlipase (DAGL) is an enzyme hydrolyzing DAG toyield 2-AG. 2-AG is transported into neurons–thetransporter is still unknown–and subsequently in-activated by the enzyme monoacylglycerol (MGL)(8, 56, 63). 2-AG synthesis and release can also bedriven by activation by Gq-coupled receptors, suchas group I mGluRs and muscarinic receptors, be-cause the signaling pathway they activate leads toaccumulation of the 2-AG precursor DAG (2, 11,20, 25).

In summary, the synthesis and release of thesetwo endogenous cannabinoids occurs on demandin an activity-dependent manner either in terms offiring of neurons or activation of Gq-coupled re-ceptors, or a combination of the two. The separatesynthesis and degradation pathways of anandam-ide and 2-AG offer an initial possibility of deter-mining their distribution and defining their roles incontrolling CNS function.

Cannabinoid Receptors: A Coupleor More

So far, two cannabinoid receptors (CB1 and CB2)have been characterized pharmacologically, ana-tomically, and by molecular cloning, but othercannabinoid receptors may exist (43). CB1 recep-tors are expressed virtually throughout the CNSfrom cortical to spinal cord regions (21). They arepredominantly localized on presynaptic terminals,

but there are reports of a postsynaptic localizationon dendrites and somata of neurons (46). Thesereceptors were initially thought to be coupled pref-erentially to a Gi/o G protein, but recent data showthat they can also couple to Gq G-protein to inducerelease of Ca2� from intracellular stores. The exis-tence of CB2 receptors has been reported in somespecific regions such as the brain stem (67), buttheir expression level is much lower than that ofCB1. They are also primarily coupled to Gi/o G-protein, but their function in the CNS is not welldefined (43). Recent data suggest that additionalcannabinoid receptors may be present in the CNSbased on a pharmacological characterization. Fi-nally, the GPR55 receptor, first identified as anorphan receptor, has also been suggested to act asa cannabinoid receptor with a signaling profile dis-tinct from CB1 and CB2 receptors (43).

Thus cannabinoid signaling in the brain isthought to be mediated primarily by CB1 recep-tors, but additional receptors may also be present.In addition, the anatomical localization of the re-ceptors in relation to the site of synthesis andrelease of endocannabinoids will determine thedirection of their signaling.

Endocannabinoids as a New Playerin Synaptic Plasticity

Activity-dependent changes in synaptic efficacyplay a critical role in shaping the functional archi-tecture of neural circuits and determining theiroperational range. In this regard, endocannabi-noids have attracted much attention in recentyears because of their unconventional way of reg-ulating synaptic transmission. There were initially

FIGURE 1. Main pathways of synthesis and degradationMain pathways of synthesis and degradation of the endocannabinoidsanandamide and 2-arachidonoylglycerol (2-AG).

REVIEWS

PHYSIOLOGY • Volume 25 • August 2010 • www.physiologyonline.org 231

by 10.220.33.2 on October 27, 2017


ownloaded from


several reports showing that activation of CB1 re-ceptors by agonists modulates synaptic transmis-sion via presynaptic mechanisms (38). A turningpoint in endocannabinoid research was the dem-onstration that they act as retrograde messengersto mediate a previously unexplained form of short-term plasticity reported in cerebellum and hip-pocampus. A brief depolarization of principalneurons (Purkinje cells or pyramidal cells) sup-presses GABAergic synaptic inputs over a few sec-onds (41, 57). This phenomenon was termeddepolarization-induced suppression of inhibition(DSI). In addition, endocannabinoid retrogradesignaling was simultaneously reported to mediatea similar suppression of excitation (DSE). The de-polarization induces an increase in the intracellu-lar Ca2� levels in the postsynaptic neurons thatmediates synthesis and release of endocannabi-noids that act on CB1 receptors on presynapticaxons and inhibit synaptic release (27, 55, 69). Sub-sequently, endocannabinoids have also beenshown to mediate long-term plasticity of both ex-citatory and inhibitory synaptic transmission inmany regions of the CNS (2, 3, 13, 17, 20, 42).

The importance of endocannabinoids for short-and long-term synaptic plasticity and the underly-ing mechanisms have been thoroughly reviewed inrecent years. In this review, we will instead focuson the role of endocannabinoid-mediated synapticplasticity in motor systems using examples fromCNS regions involved in motor behavior.

Endocannabinoids Within theSpinal Locomotor Circuitry

One region of the CNS where the link betweenendocannabinoid-mediated synaptic plasticity canbe directly related to circuit function and motorbehavior is the spinal cord. Locomotor movementsare generated by spinal networks comprised pri-marily of interconnected populations of excitatoryglutamatergic and inhibitory glycinergic interneu-rons. The motor output is generated by bursts ofmotoneuron firing that leads to the temporally se-quenced muscle contractions underlying locomo-tion. This rhythmic locomotor pattern can beproduced by the isolated spinal cord, while synap-tic transmission from excitatory and inhibitory in-terneurons can be assessed (FIGURE 2). Using thelamprey spinal cord preparation in vitro, we havebeen able to link the effects of endocannabinoid-mediated plasticity on synaptic transmission to themodulation of the locomotor circuit operation.

Spinal neurons contain the necessary machineryfor endocannabinoid signaling. For example, neu-rons in the dorsal horn express DAG lipase that isthe synthesis enzyme for the endocannabinoid2-AG. This enzyme colocalizes with group I

mGluRs (mGluR5) postsynaptically, whereas CB1receptors are found presynaptically on axon termi-nals (54). It was first shown in the cat spinal cordthat �9-tetrahydrocannabinol (THC; the activecomponent of cannabis) changes synaptic trans-mission onto motoneurons (66). These data indi-cate that cannabinoid receptors exist in the spinalcord and raise questions regarding the origin of theendocannabinoid activating them and how it af-fects the locomotor output. In recent years, the roleof endocannabinoids within the spinal locomotorcircuitry has begun to be clarified (11, 25, 33).

Endocannabinoids play an important role in set-ting the baseline locomotor frequency in the iso-lated spinal cord in vitro (25, 33). This wasdemonstrated by first inducing the locomotorrhythm with bath application of NMDA while re-cording the motor pattern in opposing ventralroots of one segment. Blockade of CB1 receptorsusing a specific antagonist reduced the baselinefrequency of the locomotor rhythm by �50%,showing that endocannabinoids are releasedwithin the locomotor circuitry and that they con-tribute to the expression of the motor pattern. Re-lease of endocannabinoids in the lamprey spinalcord can be induced on activation of mGluR1 ortachykinin receptors by substance P (25, 33, 65).

What mechanisms do endocannabinoids use toinfluence the locomotor frequency? To address thisquestion, we have examined the effect of blockingCB1 receptors on inhibitory and excitatory synap-tic transmission during locomotion. Commissuralinterneurons mediate the reciprocal midcycle in-hibition that ensures the left-right alternation ofactivity during locomotion, whereas the excitatoryinterneurons produce the on-cycle depolarizationthat drives motor activity. Blockade of CB1 recep-tors increased the amplitude of the midcycle inhi-bition, whereas it decreased that of on-cycleexcitation. This indicates that activation of CB1receptors by endocannabinoids released during lo-comotion regulates inhibitory and excitatorysynaptic transmission differently, in such a man-ner as to increase the excitability in the spinalcircuitry and thus accelerate the locomotor be-havior (11, 24, 33).

On Demand Release ofEndocannabinoids and Plasticity inthe Spinal Cord

In the lamprey spinal cord, activation of mGluR1occurs during fictive locomotion. It induces short-and long-term potentiation of the locomotor fre-quency (10, 31–33, 51). The long-term potentiationrequires activation of CB1 receptors by endocan-nabinoids release on activation of mGluR1, andblockade of CB1 receptors completely suppresses

REVIEWS

PHYSIOLOGY • Volume 25 • August 2010 • www.physiologyonline.org232

by 10.220.33.2 on October 27, 2017


ownloaded from


the long-term potentiation of the locomotor fre-quency (FIGURE 3). This long-term plasticity is me-diated by a network of modulatory signalinginvolving mGluR1 and the release of endocannabi-noids (33). The signaling pathways activated bymGluR1 lead to formation of DAG, which is a nec-essary precursor for the endocannabinoid 2-AG(23, 24, 33, 50, 51). Endocannabinoids are releasedfrom motoneurons and network interneurons “ondemand” during locomotor network operation byactivation of mGluR1 by glutamate released fromexcitatory CPG interneurons. The endocannabi-noids then act as retrograde messengers to inducelong-term depression of midcycle inhibition andlong-term potentiation of on-cycle excitation. Theendocannabinoid-mediated long-term plasticity ofsynaptic transmission and network activity alsoinvolves release of NO (FIGURE 3). Indeed, previ-ous studies in the xenopus tadpole spinal locomo-tor network have shown that NO plays a criticalrole in modulating the locomotor activity and mid-cycle inhibition (48, 49, 60). We have recentlyshown that NO is released endogenously in thelamprey spinal cord and also contributes to settingthe frequency of the locomotor rhythm (34). Sim-ilar to the effects of CB1 receptors, NO increasesthe locomotor frequency by changing the balancebetween excitatory and inhibitory synaptic trans-mission from network interneurons (34). In the

spinal circuitry, endocannabinoids and NO usesimilar synaptic mechanisms to regulate the loco-motor activity. In addition, they are recruited byactivation of mGluR1 to regulate the activity of thelocomotor network. It thus appears that endocan-nabinoids and NO signaling act synergistically tomediate long-term plasticity in the spinal circuitry;the precise mechanisms of this interaction havenot yet been clarified.

Endocannabinoids Maintain theBalance of the Basal GangliaOutput and Motor Function

The basal ganglia are thought to be responsible forthe selection of appropriate motor behavior. Theyconsist of a set of interconnected nuclei with thestriatum as the primary input nucleus receivingexcitatory inputs from cortex and thalamus and adense dopaminergic innervation from midbrainnuclei. The vast majority of neurons in the striatumare GABAergic medium spiny neurons (MSNs),with a few cholinergic and other GABAergic inter-neurons (30, 64). The basal ganglia consist of twopathways involving two distinct populations ofMSNs. Striatonigral MSNs expressing D1 receptorsproject directly to the output nuclei (the internalglobus pallidus and substantia nigra reticulata),whereas striatopallidal MSNs express D2 receptors

FIGURE 2. Levels of analysis of neural circuits controlling motor behaviorA defined motor behavior is generated by networks of excitatory and inhibitory neurons. The activity of the constituentneurons and their synaptic interactions is continuously modulated by G-protein-coupled receptors such as those acti-vated by endocannabinoids.

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


and project to output nuclei indirectly via the ex-ternal globus pallidus and the subthalamic nu-cleus. These two pathways act in opposing way tocontrol movement, with the direct one responsiblefor initiation of motor programs and the indirectone inhibiting them. Inputs to MSNs undergo syn-aptic plasticity that is thought to play a major rolein shaping the striatal output and hence initiationand termination of motor activities.

The excitatory input to MSNs displays a strongactivity-dependent plasticity in the form of long-term depression (LTD) and potentiation (LTP). Thestriatal LTD requires the elevation of Ca2� levels inpostsynaptic neurons and a convergence of mod-ulatory inputs activating group I mGluRs and D2receptors (30, 64). This combination of several sig-nals leads to release of endocannabinoids fromMSNs that act retrogradely to depress synaptictransmission from excitatory terminals. Evidence

suggests that endocannabinoids released fromMSNs is anandamide because blockade of itstransporter facilitated striatal LTD (1, 14). Finally,the LTD was blocked by D2 receptor antagonistsand enhanced by D2 agonists (30, 64). The appar-ent regulation of this endocannabinoid-mediatedLTD by D2 receptors suggests that it is selectivelyrestricted to excitatory inputs on MSNs of the indi-rect pathway. By comparing LTD in MSNs from thedirect and indirect pathways, it was shown that onlythose of the indirect pathways express LTD mediatedby endocannabinoids. D2 receptor activation en-hances this LTD by potentiating endocannabinoidsignaling (29, 30). However, endocannabinoids havebeen reported to mediate a form of LTD in MSNs ofthe direct pathway that does not depend on D2receptor activation (59, 64). The expression of LTDin MSNs of the direct pathway is inhibited by D1receptor activation, which instead induces LTP of

FIGURE 3. Endocannabinoid-mediated long-term plasticity of the spinal locomotor networkEndocannabinoids are released in the spinal cord and set the baseline locomotor burst frequency. Their release isalso triggered by activation of metabotropic glutamate receptor 1 (mGluR1). This receptor type activates Gq proteinsand phospholipase C (PLC) that hydrolyze phosphatidylinositol 4,5-bisphosphate (PIP2) to diacylglycerol (DAG) andinositol triphosphate (IP3). DAG is, in turn, hydrolyzed by DAG lipase to the endocannabinoid 2-AG. 2-AG acts as aretrograde messenger-induced long-term depression of inhibition and long-term potentiation of excitation. Nitric ox-ide (NO) is also released in the spinal cord and modulates inhibition and excitation. Endocannabinoids and NO seemto act synergistically to mediate the synaptic and network plasticity in the spinal cord.

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


excitatory synaptic transmission onto MSNs of thispathway (59). It thus appears that dopamine shiftsthe excitability balance between the direct and in-direct pathway in favor of an increased activity ofMSNs in the direct pathway, which promotes motoractivity. In this scheme, the loss of dopamine inner-vation, as in the case of Parkinson’s disease, candisrupt this balance, leading to increased excitabilityof MSNs of the indirect pathway that inhibit move-ment. To test whether the D2-dependent, endocan-nabinoid-induced LTD in striatum can restoremotor activity in the absence of dopamine inner-vation, co-administration of endocannabinoiddegradation inhibitors and D2 agonists was per-formed (29, 30). As predicted, this combined treat-ment in animals with depleted dopaminergicinnervation significantly enhanced locomotor ac-tivity. These results provide evidence that endocan-nabinoid-mediated synaptic plasticity in the striatumcan be important in controlling motor function.However, it should be noted that endocannabinoidsignaling, which is the topic of this review, cannotalone be responsible for balancing the activity of thedirect and indirect pathways because other mecha-nisms are also necessary for the normal function ofthe basal ganglia circuitry and selection of motorbehavior (64).

Cerebellar Function: A Rolefor Endocannabinoids inMotor Learning

The cerebellum plays an important role in fine-tuning of motor behavior and in learning of newmotor tasks. Purkinje cells are the only output fromthe cerebellar cortex and project to the deep cere-bellar nuclei. Each Purkinje cell receives excitatoryinputs from many parallel fibers arising from gran-ule cells and from a single climbing fiber originat-ing in the inferior olive. Excitatory synaptictransmission to Purkinje cells displays both short-and long-term changes that depend on release ofendocannabinoids and activation of CB1 recep-tors. Endocannabinoids have been shown to me-diate depolarization-mediated suppression of bothinhibitory and excitatory synaptic transmission toPurkinje cells (9, 27, 28, 70). The suppression ofexcitatory transmission can also be induced bysynaptic activation. It was shown that high-fre-quency stimulation of parallel fibers can lead toactivation of mGluR1, resulting in endocannabi-noid synthesis via the PLC-DAG pathway (44, 45).This synaptic suppression of excitation is blockedby CB1 receptor antagonists and is absent in CB1receptor knockout mice, confirming a role of en-docannabinoid release in this form of synapticplasticity (3, 20).

As in basal ganglia, LTD is also a prominent formof plasticity in the cerebellum (6, 19, 37). Coinci-dent activation of parallel and climbing fibers overtime leads to weakening of parallel fiber synapsesonto Purkinje cells. This LTD has been proposed tomediate motor learning in the vestibulo-ocular re-flex pathway (18), whereby climbing fibers signalmotor error and weaken simultaneously active par-allel fiber synapses. This results in inhibition of theincorrect movement and an improvement of mo-tor performance. There is a consensus that LTD ismediated postsynaptically, although the underly-ing mechanisms and its significance for motor per-formance has long been debated over several years(19). A recent contribution to this debate has beenprovided by evidence that LTD depends on therelease of endocannabinoids from Purkinje cells(58). The question that arises is how can retrogradesignaling by endocannabinoids mediate LTD thatis expressed postsynaptically? One possible answerinvolves nitric oxide (NO), which has been sug-gested to be released from presynaptic terminalson activation of CB1 receptors and to act as ananterograde messenger (58). Further studies, how-ever, are needed to determine precisely how aninteraction between NO and endocannabinoid sys-tems mediated LTD in the cerebellum.

The cerebellum contributes to the precision andsmooth execution of motor tasks. Although endo-cannabinoid-mediated plasticity has been studiedextensively in the cerebellum, very little is knownabout its significance for cerebellar function andmotor behavior. Mice lacking CB1 receptors arenot ataxic and do not display aberrant motor con-trol (36, 72). To understand the function of endo-cannabinoid-mediated plasticity in terms of motorfunction, the analysis needs to be broadened fromsingle synapses to circuit operation and ultimatelyto behaviorally relevant tasks.

Role of Endocannabinoids inStabilizing Motor Map Formationand Maintenance

Specific regions of the neocortex, particularly thesupplementary motor area, are important for plan-ning of movement and execution of fine motortasks. There is a continuous interaction betweensensory and motor areas to integrate incomingsensory inputs and transform them into appropri-ate motor behavior. The organization of sensorymaps not only defines representation of specificbody regions in the cortex but may also be criticalfor the organization of the cortical motor circuits.We will now briefly review the role of endocan-nabinoids in synaptic plasticity in the neocortexand their significance in circuit plasticity.

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


Endocannabinoids have been shown to play acritical role in long-term depression (LTD) in theneocortex (61). Pairing of presynaptic and postsyn-aptic activity can trigger long-term potentiation(LTP) or depression depending on their relativetiming. LTD occurs at many neocortical synapsesand is elicited when postsynaptic firing precedespresynaptic firing (5). Postsynaptic calcium eleva-tion and activation of group I mGluRs drivepostsynaptic endocannabinoid synthesis and re-lease, which signals retrogradely to presynapticCB1 receptors, driving a long-lasting decrease inrelease probability (12, 17, 52, 61). This LTD alsorequires activation of presynaptic NMDA recep-tors. In this scheme, presynaptic CB1 receptorsdetect postsynaptic activity as a consequence ofrelease of endocannabinoids, whereas presynapticNMDA receptors are a sensor of presynaptic firingsince they act as autoreceptors. Because endocan-nabinoids can diffuse to other synapses, the coin-cidence of activation of presynaptic CB1 andNMDA receptors could increase the specificity ofthis diffusible retrograde signal to restrict the plas-ticity to the active synapses.

Does endocannabinoid-mediated LTD occurin vivo, and what role does it play in motor control?In recent years, endocannabinoid-induced LTDhas been suggested to contribute to depression ofexcitatory synaptic transmission in barrel (S1) andvisual (V1) cortex. Whisker deprivation results inLTD of excitatory synaptic transmission in de-prived columns in the barrel cortex that resemblesendocannabinoid-induced LTD in vitro (12). Thesensory deprivation may result in weakening ofdeprived whisker representation as a result of LTDof synaptic interactions by endocannabinoids. In arecent study, Li and colleagues (39) showed thatsignaling via CB1 receptors is required for whiskermap development. These authors argue that endo-cannabinoid-mediated LTD may act to weakeninappropriate synapses and contribute to activity-dependent organization of sharp whisker borders(maps). Similarly, monocular deprivation de-presses visually evoked responses, an effect alsothought to be mediated by CB1 receptors (4). Asystemic pharmacological blockade of CB1 recep-tors in vivo prevents depression of closed-eye re-sponses, suggesting that CB1-LTD is a criticalmechanism for response depression (40). Althoughendocannabinoid-mediated LTD has been shownto be involved in the organization of sensory maps,it is not the only mechanism involved, and othermechanisms also play a role (12, 62).

These two examples show that endocannabinoidretrograde signaling can shape map representationin the neocortex. Sensory signal processing in dif-ferent cortical areas needs ultimately to be trans-formed into motor action, for example during

visuo-motor coordination or exploratory move-ments. The endocannabinoid-mediated synapticplasticity may thus play a role in refining sensory-motor circuit organization and adapting theirrange of operation during motor performance andduring learning of new motor tasks.

Balancing Motor Functionsby Endocannabinoids

From the example discussed above, it is clear thatendocannabinoids play an important role in bal-ancing the excitability of circuits controlling motorbehavior. The long-term plasticity induced by en-docannabinoids may provide a long-term bufferthat stabilizes the organization of motor circuitsand their activity. In the spinal cord, the shift in theexcitability balance induced by endocannabinoidsdoes not only lead to an increase in the locomotorfrequency, but it also primes the circuitry to fasteronset of locomotion with a lesser excitatory drivefrom descending command centers. The endocan-nabinoid-dependent plasticity in critical motor ar-eas in the brain plays a role in refining andconsolidating sensory-motor maps during learningof new motor task. In addition, by tuning synaptictransmission, this novel signaling mechanisms canprovide a long-term buffer that stabilizes the deci-sion-making process, initiation, and the control ofprecision of movement, thus permitting the motorprograms to be effortlessly and unconsciouslyexecuted.

Perspectives

Like the motor systems they modulate, CB recep-tors and endocannabinoid signaling are distrib-uted throughout the rostrocaudal neuraxis. Thisraises the possibility that 1) there are features thatare common to the different systems in terms ofendocannabinoid signaling, and 2) the spinal cordrole could, for phylogenetic and developmental rea-sons, represent the primordial/ancient condition.

The possibility of on demand release of endo-cannabinoids from network neurons that are ac-tive during locomotion alters our understanding ofmodulatory systems. Initially, modulation wasthought to arise from sets of dedicated neurons,with the modulatory transmitter being releasedfrom axon terminals. The ability of network neu-rons to release endocannabinoids from their den-drites and soma shows that every neuron,including motoneurons, can be transformed into amodulatory neuron in an activity-dependent man-ner and thereby set the strength of the excitatoryand inhibitory synaptic transmission it receives.This adds to the dynamic processing taking placewithin the spinal circuitry to generate and regulate

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


the locomotor pattern. Retrograde signaling by en-docannabinoids can be considered an integral partof the locomotor pattern generation. Enhancingthis signaling may help to promote recovery offunction after spinal cord injury.

Motor behavior is orchestrated by interactionsbetween many transmitters and modulatory sys-tems. The final action of endocannabinoids doesnot always solely involve activation of CB recep-tors, but they can also interact with other signalingsystems. In the cerebellum and in the spinal loco-motor network, endocannabinoids interact in asynergistic manner with NO signaling to mediatesynaptic and network plasticity. The interplay be-tween these two unconventional signaling mole-cules suggests that the activity of a given networkunderlying motor behavior is not only dependenton synaptic connectivity but is also continuouslyregulated by a network of modulatory systems.Thus an understanding of how the CNS generatesmotor behavior will require defining the connec-tivity of both the neural circuit and the biochemi-cal modulatory networks. �

We thank Drs. Sten Grillner, Russell Hill, Gilad Silber-berg, and Keith Sillar for comments on the manuscript.

The work in the authors’ laboratory is supported by theSwedish Research Council, Söderberg Foundation, the Eu-ropean Commission (Health-F2-2007-201144), and theKarolinska Institutet.

No conflicts of interest, financial or otherwise, are de-clared by the author(s).

References1. Ade KK, Lovinger DM. Anandamide regulates postnatal de-

velopment of long-term synaptic plasticity in the rat dorso-lateral striatum. J Neurosci 27: 2403–2409, 2007.

2. Alger BE. Retrograde signaling in the regulation of synaptictransmission: focus on endocannabinoids. Prog Neurobiol68: 247–286, 2002.

3. Chevaleyre V, Takahashi KA, Castillo PE. Endocannabinoid-mediated synaptic plasticity in the CNS. Annu Rev Neurosci29: 37–76, 2006.

4. Crozier RA, Wang Y, Liu CH, Bear MF. Deprivation-inducedsynaptic depression by distinct mechanisms in different lay-ers of mouse visual cortex. Proc Natl Acad Sci USA 104:1383–1388, 2007.

5. Dan Y, Poo MM. Spike timing-dependent plasticity: fromsynapse to perception. Physiol Rev 86: 1033–1048, 2006.

6. Daniel H, Levenes C, Crepel F. Cellular mechanisms of cere-bellar LTD. Trends Neurosci 21: 401–407, 1998.

7. Destexhe A, Marder E. Plasticity in single neuron and circuitcomputations. Nature 431: 789–795, 2004.

8. Di Marzo V. Endocannabinoids: synthesis and degradation.Rev Physiol Biochem Pharmacol 160: 1–24, 2008.

9. Diana MA, Levenes C, Mackie K, Marty A. Short-term retro-grade inhibition of GABAergic synaptic currents in rat Pur-kinje cells is mediated by endogenous cannabinoids. JNeurosci 22: 200–208, 2002.

10. El Manira A, Kettunen P, Hess D, Krieger P. Metabotropicglutamate receptors provide intrinsic modulation of the lam-prey locomotor network. Brain Res Rev 40: 9–18, 2002.

11. El Manira A, Kyriakatos A, Nanou E, Mahmood R. Endocan-nabinoid signaling in the spinal locomotor circuitry. Brain ResRev 57: 29–36, 2008.

12. Feldman DE. Synaptic mechanisms for plasticity in neocortex.Annu Rev Neurosci 32: 33–55, 2009.

13. Freund TF, Katona I, Piomelli D. Role of endogenous canna-binoids in synaptic signaling. Physiol Rev 83: 1017–1066,2003.

14. Gerdeman GL, Ronesi J, Lovinger DM. Postsynaptic endo-cannabinoid release is critical to long-term depression in thestriatum. Nat Neurosci 5: 446–451, 2002.

15. Grillner S. The motor infrastructure: from ion channels toneuronal networks. Nat Rev Neurosci 4: 573–586, 2003.

16. Grillner S. Biological pattern generation: the cellular andcomputational logic of networks in motion. Neuron 52: 751–766, 2006.

17. Heifets BD, Castillo PE. Endocannabinoid signaling and long-term synaptic plasticity. Annu Rev Physiol 71: 283–306, 2009.

18. Ito M. Neural design of the cerebellar motor control system.Brain Res 40: 81–84, 1972.

19. Ito M. Cerebellar long-term depression: characterization, sig-nal transduction, and functional roles. Physiol Rev 81: 1143–1195, 2001.

20. Kano M, Ohno-Shosaku T, Hashimotodani Y, Uchigashima M,Watanabe M. Endocannabinoid-mediated control of synaptictransmission. Physiol Rev 89: 309–380, 2009.

21. Katona I, Freund TF. Endocannabinoid signaling as a synapticcircuit breaker in neurological disease. Nat Med 14: 923–930,2008.

22. Katz PS, Frost WN. Intrinsic neuromodulation: altering neu-ronal circuits from within. Trends Neurosci 19: 54–61, 1996.

23. Kettunen P, Hess D, El Manira A. mGluR1, but not mGluR5,mediates depolarization of spinal cord neurons by blocking aleak current. J Neurophysiol 90: 2341–2348, 2003.

24. Kettunen P, Krieger P, Hess D, El Manira A. Signaling mech-anisms of metabotropic glutamate receptor 5 subtype and itsendogenous role in a locomotor network. J Neurosci 22:1868–1873, 2002.

25. Kettunen P, Kyriakatos A, Hallen K, El Manira A. Neuromodu-lation via conditional release of endocannabinoids in thespinal locomotor network. Neuron 45: 95–104, 2005.

26. Kiehn O. Locomotor circuits in the mammalian spinal cord.Annu Rev Neurosci 29: 279–306, 2006.

27. Kreitzer AC, Regehr WG. Retrograde inhibition of presynap-tic calcium influx by endogenous cannabinoids at excitatorysynapses onto Purkinje cells. Neuron 29: 717–727, 2001.

28. Kreitzer AC, Regehr WG. Cerebellar depolarization-inducedsuppression of inhibition is mediated by endogenous canna-binoids. J Neurosci 21: RC174, 2001.

29. Kreitzer AC, Malenka RC. Endocannabinoid-mediated rescueof striatal LTD and motor deficits in Parkinson’s disease mod-els. Nature 445: 643–647, 2007.

30. Kreitzer AC, Malenka RC. Striatal plasticity and basal gangliacircuit function. Neuron 60: 543–554, 2008.

31. Krieger P, Grillner S, El Manira A. Endogenous activation ofmetabotropic glutamate receptors contributes to burst fre-quency regulation in the lamprey locomotor network. Eur JNeurosci 10: 3333–3342, 1998.

32. Krieger P, Hellgren-Kotaleski J, Kettunen P, El Manira AJ.Interaction between metabotropic and ionotropic glutamatereceptors regulates neuronal network activity. J Neurosci 20:5382–5391, 2000.

33. Kyriakatos A, El Manira A. Long-term plasticity of the spinallocomotor circuitry mediated by endocannabinoid and nitricoxide signaling. J Neurosci 27: 12664–12674, 2007.

34. Kyriakatos A, Molinari M, Mahmood R, Grillner S, Sillar KT, ElManira A. Nitric oxide potentiation of locomotor activity inthe spinal cord of the lamprey. J Neurosci 29: 13283–13291,2009.

35. LeBeau FE, El Manira A, Griller S. Tuning the network: mod-ulation of neuronal microcircuits in the spinal cord and hip-pocampus. Trends Neurosci 28: 552–561, 2005.

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


36. Ledent C, Valverde O, Cossu G, Petitet F, AubertJF, Beslot F, Bohme GA, Imperato A, PedrazziniT, Roques BP, Vassart G, Fratta W, Parmentier M.Unresponsiveness to cannabinoids and reducedaddictive effects of opiates in CB1 receptorknockout mice. Science 283: 401–404, 1999.

37. Levenes C, Daniel H, Crepel F. Long-term de-pression of synaptic transmission in the cerebel-lum: cellular and molecular mechanisms revisited.Prog Neurobiol 55: 79–91, 1998.

38. Levenes C, Daniel H, Soubrie P, Crepel F. Can-nabinoids decrease excitatory synaptic transmis-sion and impair long-term depression in ratcerebellar Purkinje cells. J Physiol 510: 867–879,1998.

39. Li L, Bender KJ, Drew PJ, Jadhav SP, SylwestrakE, Feldman DE. Endocannabinoid signaling is re-quired for development and critical period plas-ticity of the whisker map in somatosensorycortex. Neuron 64: 537–549, 2009.

40. Liu CH, Heynen AJ, Shuler MG, Bear MF. Canna-binoid receptor blockade reveals parallel plastic-ity mechanisms in different layers of mouse visualcortex. Neuron 58: 340–345, 2008.

41. Llano I, Leresche N, Marty A. Calcium entry in-creases the sensitivity of cerebellar Purkinje cellsto applied GABA and decreases inhibitory syn-aptic currents. Neuron 6: 565–574, 1991.

42. Lovinger DM. Presynaptic modulation by endo-cannabinoids. Handb Exp Pharmacol: 435–477,2008.

43. Mackie K, Stella N. Cannabinoid receptors andendocannabinoids: evidence for new players.AAPS J 8: E298–E306, 2006.

44. Maejima T, Ohno-Shosaku T, Kano M. Endoge-nous cannabinoid as a retrograde messengerfrom depolarized postsynaptic neurons to pre-synaptic terminals. Neurosci Res 40: 205–210,2001.

45. Maejima T, Oka S, Hashimotodani Y, Ohno-Shosaku T, Aiba A, Wu D, Waku K, Sugiura T,Kano M. Synaptically driven endocannabinoid re-lease requires Ca2�-assisted metabotropic gluta-mate receptor subtype 1 to phospholipaseCbeta4 signaling cascade in the cerebellum. JNeurosci 25: 6826–6835, 2005.

46. Marsicano G, Goodenough S, Monory K, Her-mann H, Eder M, Cannich A, Azad SC, CascioMG, Gutierrez SO, van der Stelt M, Lopez-Rodriguez ML, Casanova E, Schutz G, Zieglgan-sberger W, Di Marzo V, Behl C, Lutz B. CB1cannabinoid receptors and on-demand defenseagainst excitotoxicity. Science 302: 84–88, 2003.

47. McKinney MK, Cravatt BF. Structure and functionof fatty acid amide hydrolase. Annu Rev Biochem74: 411–432, 2005.

48. McLean DL, Sillar KT. Nitric oxide selectivelytunes inhibitory synapses to modulate vertebratelocomotion. J Neurosci 22: 4175–4184, 2002.

49. McLean DL, Sillar KT. Metamodulation of a spinallocomotor network by nitric oxide. J Neurosci 24:9561–9571, 2004.

50. Nanou E, El Manira A. Mechanisms of modulationof AMPA-induced Na�-activated K� current bymGluR1. J Neurophysiol 103: 441–445, 2010.

51. Nanou E, Kyriakatos A, Kettunen P, El Manira A.Separate signalling mechanisms underlie mGluR1modulation of leak channels and NMDA recep-tors in the network underlying locomotion. JPhysiol 587: 3001–3008, 2009.

52. Nevian T, Sakmann B. Spine Ca2� signaling inspike-timing-dependent plasticity. J Neurosci 26:11001–11013, 2006.

53. Nusbaum MP, Blitz DM, Swensen AM, Wood D,Marder E. The roles of co-transmission in neuralnetwork modulation. Trends Neurosci 24: 146–154, 2001.

54. Nyilas R, Gregg LC, Mackie K, Watanabe M, Zim-mer A, Hohmann AG, Katona I. Molecular archi-tecture of endocannabinoid signaling atnociceptive synapses mediating analgesia. Eur JNeurosci 29: 1964–1978, 2009.

55. Ohno-Shosaku T, Maejima T, Kano M. Endoge-nous cannabinoids mediate retrograde signalsfrom depolarized postsynaptic neurons to pre-synaptic terminals. Neuron 29: 729–738, 2001.

56. Piomelli D. The molecular logic of endocannabi-noid signalling. Nat Rev Neurosci 4: 873–884,2003.

57. Pitler TA, Alger BE. Postsynaptic spike firing re-duces synaptic GABAA responses in hippocam-pal pyramidal cells. J Neurosci 12: 4122–4132,1992.

58. Safo PK, Regehr WG. Endocannabinoids controlthe induction of cerebellar LTD. Neuron 48: 647–659, 2005.

59. Shen W, Flajolet M, Greengard P, Surmeier DJ.Dichotomous dopaminergic control of striatalsynaptic plasticity. Science 321: 848–851, 2008.

60. Sillar KT, Combes D, Ramanathan S, Molinari M,Simmers J. Neuromodulation and developmentalplasticity in the locomotor system of anuran am-phibians during metamorphosis. Brain Res Rev57: 94–102, 2008.

61. Sjostrom PJ, Turrigiano GG, Nelson SB. Neocor-tical LTD via coincident activation of presynapticNMDA and cannabinoid receptors. Neuron 39:641–654, 2003.

62. Smith GB, Heynen AJ, Bear MF. Bidirectionalsynaptic mechanisms of ocular dominance plas-ticity in visual cortex. Philos Trans R Soc Lond BBiol Sci 364: 357–367, 2009.

63. Sugiura T, Kobayashi Y, Oka S, Waku K. Biosyn-thesis and degradation of anandamide and2-arachidonoylglycerol and their possible physi-ological significance. Prostaglandins Leukot Es-sent Fatty Acids 66: 173–192, 2002.

64. Surmeier DJ, Plotkin J, Shen W. Dopamine andsynaptic plasticity in dorsal striatal circuits con-trolling action selection. Curr Opin Neurobiol 19:621–628, 2009.

65. Thorn Perez C, Hill RH, El Manira A, Grillner S.Endocannabinoids mediate tachykinin-inducedeffects in the lamprey locomotor network. J Neu-rophysiol 102: 1358–1365, 2009.

66. Turkanis SA, Karler R. Effects of delta 9-tetrahy-drocannabinol on cat spinal motoneurons. BrainRes 288: 283–287, 1983.

67. Van Sickle MD, Duncan M, Kingsley PJ, MouihateA, Urbani P, Mackie K, Stella N, Makriyannis A,Piomelli D, Davison JS, Marnett LJ, Di Marzo V,Pittman QJ, Patel KD, Sharkey KA. Identificationand functional characterization of brainstem can-nabinoid CB2 receptors. Science 310: 329–332,2005.

68. Wang J, Ueda N. Biology of endocannabinoidsynthesis system. Prostaglandins Other Lipid Me-diat 89: 112–119, 2009.

69. Wilson RI, Nicoll RA. Endogenous cannabinoidsmediate retrograde signalling at hippocampalsynapses. Nature 410: 588–592, 2001.

70. Yoshida T, Hashimoto K, Zimmer A, Maejima T,Araishi K, Kano M. The cannabinoid CB1 receptormediates retrograde signals for depolarization-induced suppression of inhibition in cerebellarPurkinje cells. J Neurosci 22: 1690–1697, 2002.

71. Zhang W, Pombal MA, el Manira A, Grillner S.Rostrocaudal distribution of 5-HT innervation inthe lamprey spinal cord and differential effects of5-HT on fictive locomotion. J Comp Neurol 374:278–290, 1996.

72. Zimmer A, Zimmer AM, Hohmann AG, Herken-ham M, Bonner TI. Increased mortality, hypoac-tivity, and hypoalgesia in cannabinoid CB1receptor knockout mice. Proc Natl Acad Sci USA96: 5780–5785, 1999.

REVIEWS


by 10.220.33.2 on October 27, 2017


ownloaded from


reviews - semantic scholar...integrated central nervous system (cns) process-ing. the cns comprises...

Documents