robert c. malenka and jeffery d. kocsis- presynaptic actions of carbachol and adenosine on...

8/3/2019 Robert C. Malenka and Jeffery D. Kocsis- Presynaptic Actions of Carbachol and Adenosine on Corticostriatal Synapti

1/7

The Journal of Neuroscience, October 1988. &IO ): 3750-3756

Presynaptic Actions of Carbachol and Adenosine on CorticostriatalSynaptic Transmission Studied in vitroRobert C. Malenka and Jeffery D. Kocsis2Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, California 94305and Veterans Administration Medical Center, Palo Alto, California 94304, and 2Department of Neurology, Yale UniversityMedical School, New Haven, Connecticut 06410

The purpose of this study was to iden tify, in the in vitro ratneostriatal slice preparation, an electrophysiological re-sponse corresponding to the activation of striatal neuronsby cortical afferents, and to study the actions of a variety ofputative neurotransmitters on modulating synapt ic trans-mission at this synapse. Local stimulation of the neostriatalslice evokes a field potential composed of 2 prominent nega-tivit ies. Experiments using calcium antagonists and tetro-dotoxin indicate that the first negat ivity (Nl) reflects thedirect activation of intrinsic neurons and axons, while thesecond negativ ity (N2) is synapt ically mediated. The afferentfibers eliciting the second negat ivity have not previouslybeen identified because of the mixing of afferents within thestriatum. However, similar field potential responses are elic-ited by stimulation of the corpus callosum, in which the onlystriatal afferents are crossed corticostriatal fibers. Kynuren-ate, y-D-glutamylglyc ine, and piperidine dicarboxylate all re-vers ibly reduced or abolished N2, suggesting that the trans-mitter generating the response is an excitatory amino acid.Together, these results strongly suggest that N2 reflectsactivation of striatal neurons by corticostriatal afferents.

Three putative striatal neurotransmitters, dopamine,acetylcholine and adenosine, were studied with respect totheir ability to modulate the corticostriatal response. Do-pamine had minimal effect s on the waveform of the fieldpotential. In contrast, carbachol or adenosine consistentlyreversib ly reduced or eliminated N2. Atropine blocked car-bachols actions, while theophyl line blocked adenosines ac-tions, indicating that the compounds were acting on mus-carinic and adenosine receptors, respectively. To testwhether these were primarily pre- or postsynaptic actions,we recorded the response to ionophoretically applied glu-tamate while simultaneously recording the synapt icallyevoked field potential from the same location. Kynurenatereduced both responses simultaneously, while carbachol oradenosine reduced only the synapt ically mediated N2. Fromthese results we propose that the terminals of corticostriatalafferents may possess muscarinic and adenosine receptorsthat function to inhibit transmitter release.

Received Sept. 3, 1987; revised Jan. 22, 1988; accepted Jan. 29, 1988.Th is work wa s supported by grants from the Medical Research Service of theVeterans Administration and N.I.H.Corresponde nce shou ld be addresse d to Robert C. Malenka, M.D., Ph.D., De-partment of Psychiatry and Behaviorial Science s, Stanford University School ofMedicine, Stanford, CA 94305

Copyright 0 1988 Socie ty for Neuro science 0270-6474/88/10375 0-07$02.00/O

The mammalianneostriatum containsa variety of putative neu-rotransmitters, including the highest concentrations of acetyl-choline and dopamine ound amongany structures n the mam-malian forebrain (Graybiel and Ragsdale,1983). t thus hasbeenthe subject of much researchaimed at elucidating its biochem-ical anatomy. Electrophysiological analysis has proven moredifficult, primarily becauseof its nonlaminar organization andthe limitations inherent in utilizing in vivo preparations. Re-cently, however, several nvestigatorshave begunexamining theelectrophysiological properties of striatal neurons using the invitro striatal slice preparation (Misgeld et al., 1979; Kita et al.,1985; Cordingley and Weight, 1986).The purpose of this study was to identify, in the slice prep-aration, the electrophysiological esponse eneratedby the cor-tical input to the neostriatum and to examine the actions ofcompounds that may presynaptically inhibit this input. Theentire cerebral cortex projects topographically to the neostria-turn and thereby provides a major afferent input (Kemp andPowell, 1971). Presynaptic modulation of the corticostriatalpathway would have profound consequencesor the transfer ofinformation from cortex to striatum and for the functioning ofthe extrapyramidal, aswell asother behaviorally relevant, sys-tems.Yamamoto (1973) first reported that local stimulation in thestriatal slice evokes a field potential consisting of 2 negativewaves. It hasbeensuggestedhat the first component representsthe direct and antidromic activation of axons and intrinsic neu-rons, while the secondcomponent appears o be synapticallymediated (Takagi and Yamamoto, 1978; Misgeld et al., 1979;Cordingley and Weight, 1986). The identity of the neurotrans-mitter responsible or generating the secondnegativity is un-clear. Misgeld and his colleagues1980) have reported that nic-otinic antagonists lock this component,and thereforeconcludedthat it is mediated by ACh released rom intrinsic cholinergicneurons and acting on nicotinic receptors. On the other hand,Cordingley and Weight (1986) have found that nicotinic antag-onistshave no effect on the field potentials secondcomponentand instead have proposed that it is mediated by a glutamate-like transmitter. In the initial part of this study, we examine theactions of both nicotinic antagonistsand excitatory amino acidantagonistson the field potential evoked by local stimulationin the in vitro rat neostriatal slice preparation. In addition, bycarefully positioning stimulating and recording microelectrodes,we attempt to unequivocally identify a corticostriatal response.Several compoundsmay have significant actions on the cor-tical input to the neostriatum, specifically, dopamine DA), car-bachol, and adenosine. t hasbeen suggestedhat DA receptors


2/7

The Journal s f Neuroscience, October 1988, 8(10) 3751

* CONTROL

D KYNURENATE

B 0.5 Ca,6 Mg WASH

I3 msec

E WASH F TTX

are found on the terminals of the cortical afferents to the striatum(Schwartz et al., 1978; Komhuber and Komhuber, 1986) andthat DA can inh ibi t the release of glutamate from striatal slices(Mitchell and Doggett, 1980; Rowlands and Roberts, 1980).Muscarinic agonists, such as carbachol, have been reported todecrease the second negativ ity of the field potential, perhaps byacting at a presynaptic locus (Takagi and Yamamoto, 1978;Weiler et al., 1984). Finally, adenosineand its receptors arefound in reasonablyhigh concentrations n the striatum (Good-man and Snyder, 1982) and in a variety of other systemshavebeen shown o presynaptically inhibit excitatory synaptic trans-mission (Ginsborg and Hirst, 1972; Okada and Ozawa, 1980;Kocsis et al., 1984; Dunwiddie and Haas, 1985). In the secondpart of this study, we examine the actions of DA, carbachol,and adenosine on the corticostriatal response.Materials and MethodsNeostriatal slices were prepared from female, Wistar rats (180-240 gm)using standard techniques. The animals were deeply anesthetized withsodium pentobarbital(60 mg/kg, i.p.) and decapitated. The dorsal sur-face of the brain was immediately exposed, and a 4-6 mm coronalsection of one hemisphere was removed by hemisecting the brain andthen making coronal cuts approximately 3 and 7-9 mm from the tip ofthe frontal poles. This tissue block was immediately immersed in cold(2+C), modified Krebs solution and then glued to the stage of a Vi-bratome with cyanoacrylate. Coronal slices, 400 pm thick, were thencut, and included neocortex, corpus callosum, and neostriatum. Sliceswere stored in oxygenated modified Krebs solution at room temperatureand were allowed to stabilize for at least 1 hr before being transferredto the recording chamber. A single slice was transported to the recordingchamber (volume, 1.0 cm), where it was held between 2 nylon nets,submerged beneath continuously flowing (3-5 cm/min) modified Krebssolution consisting of (in mM) NaCl, 124; KCl, 3.0; MgCl,, 2.0; CaCl,,2.0; NaHCO,, 26.0; NaH,PO,, 1.3; dextrose, 10.0. The solution wassaturated with 95% 0, and 5% CO, (pH 7.4). All drugs were obtainedfrom Sigma and were made up at high concentrations (lo-100 mM) indistilled water and diluted to their final concentrations in the bathingmedium immediately before application.Standard f ield potential recording techniques were used. Electrodes(2-8 MO) were pulled from omega dot glass capillary tubing (2.0 mmO.D., 1.0 mm I.D.) and filled with 3 M NaCl. Signals were amplifiedusing an Axon Instruments Axoprobe- 1 and then digitized with a Ni-colet 1170 signal averager and stored on magnetic tape. Slices werestimulated by constant-current cathodal pulses (40 psec duration) de-livered through monopolar tungsten electrodes. The stimulating andrecording microelectrodes were positioned approximately 0.5-3.0 mm

Figure 1. N2 is synaptically mediatedand is blocked by kynurenate. A-C, Ap-plication for 3 min o f a solution con-taining 0.5 mM CaCl, and 6.0 mM MgCl,reversibly blocked N2 without affectingNl. C-E, Kynurenate (1 mM) also re-@Mm

versibly blocked N2 within 2 min ofapplication. F, TTX (0.3 PM) blockedboth Nl and N2. In this and all sub-sequent experiments, a stable controlfield potential was obtained and drugswere then sequentially applied to thesame slice.

apart, with the stimulating microelectrode placed either directly in thecorpus callosum or immediately (within 500 pm) ventral to it and therecording microelectrode placed ventral to the stimulating electrode.Sodium glutamate (1 .O M, pH 8) was applied ionophoretically by po-sitioning the ionophoretic microelectrode as close as possible to therecording microelectrode under visual control and then moving it throughthe depth of the slice to obtain a maximal response. At the conclusionof each experiment involving glutamate ionophoresis, the ionophoreticmicroelectrode was raised out of the slice so that its tip was in thebathing medium and the electrical response to glutamate applicationwas then recorded. The coupling art ifact thus recorded was digitallysubtracted f rom the previously recorded glutamate responses.ResultsThe field potential elicited by intrastriatal stimulation consistsof 2 negativities (Nl and N2; Fig. 1A) (Yamamoto, 1973; Mis-geld et al., 1979; Cordingley and Weight, 1986). Nl is thoughtto reflect direct activation of neuronal elements,while N2 issynaptically mediated (Takagi and Yamamoto, 1978; Misgeldet al., 1979; Cordingley and Weight, 1986). However, there issomequestion about the transmitter and pathway mediating N2(Takagi and Yamamoto, 1978; Misgeld et al., 1980; Cordingleyand Weight, 1986). Figure 1, A-C, shows hat N2 is reversiblyblocked by a low-calcium (0.5 mM), high-magnesium 6.0 mM)solution, while Nl is unaffected (n = 5). The nicotinic antago-nists mecamylamine (100 PM; n = 3) and d-tubocurarine (30PM; n = 4) had no effect on either component of the field po-tential, even when applied for up to 30 min. In contrast, kyn-urenate (1 mM), an excitatory amino acid antagonist (Ganonget al., 1983; Jahr and Yoshioka, 1986) reversibly blocked orreduced N2 within 2 min of bath application without affectingNl (n = 21; Fig. 1, C-E). Both Nl and N2 were blocked byTTX (n = 3; Fig. lF), indicating that they are both dependenton sodium-mediatedspike discharges.As would be expected, Nl was virtually unaffected by anydose of kynurenate at any stimulation strength (Fig. 2A). Theantagonistic action of kynurenate on N2 was dose-dependentand observed at all stimulation strengths Fig. 2B). To furthertest whether N2 is mediated by an excitatory amino acid, theeffects of 2 other excitatory amino acid antagonistswere ex-amined. Both y-o-glutamylglycine (1 mM; n = 3) and piperidinedicarboxylate (1 mM; y1 = 3; Fig. 6P) reversibly reduced orblocked N2 without affecting Nl.


3/7

3752 Malenka and Kocsis - Presynaptic Inhibition of Corticostr iatal Pathwa y

A Nl * CONTROL B DOPAMINE2.0 -1.6-

s.54 1.2-g5 0.8 -i

0 Control* 0.2 mMA 0.4 mM0 0.6 mMA 1.6 mMo Wash *

r I-I 0.5 mV3 msec

C WASH

d4

D KYNURENATE

N2 l 0. 0 00

lx0 xx.

0Figure 3. Dopamine as o effecton the ocallyevoked ieldpotential.A-C, Applicationof DA (30 BM ) hadno significant ffecton eitherNlor N2. All solutions ontained .3 mM ascorbate., Application ofkynurenate1 mM) o the sameieldpotentialsignificantlyeduced 2.ascorbate;Fig. 31had no consistentor significanteffectson eitherNl or N2. This lack of effect of DA on N2 is consistentwiththe resultsof 2 recent studies, which call into question the ex-istence of DA receptors on the terminals of corticostriatal af-ferents (Trugman et al., 1986; Joyce and Marshall, 1987).In contrast to DA, muscarinic agonistshave been reported tosignificantly reduce N2 and intracellularly recorded EPSPs Ta-kagi and Yamamoto, 1978; Misgeld et al., 1982; Weiler et al.,1984). n agreementwith these eports, we found that carbachol(5-20 ELM), a muscarinic agonist, reversibly reduced or blockedN2 (Figs. 4, 5; y1= 13). Carbachols action was prevented bythe coapplication of atropine (l-3 PM ) (Fig. 4E; n = 7), dem-onstrating that carbachol was ndeed acting on muscarinic re-ceptors in the striatum. Figure 4F also demonstrates hat thesameN2 component that was reduced by carbachol was alsoeliminated by kynurenate. Contrary to previous reports (Mis-geld et al., 1982; Weiler et al., 1984),application of atropine byitself did not increase he amplitude of N2 (Fig. 40), even whenthe N2 amplitude was clearly submaximal.To begin to test whether carbachol was exerting its actionspre- or postsynaptically, the response o ionophoretically ap-plied glutamate was recorded while simultaneously recordingthe synaptically evoked field potential from the same ocation.If an agent s acting presynaptically and inhibiting transmitterrelease, hen the ionophoretically induced glutamate field po-tential should be unaffected at a time when the synapticallymediated ield is reduced.An agentacting postsynapticallywouldbe expected to affect both field potentials to a similar degree(seeDiscussion or possibleproblems with this interpretation).

Figure 5 shows hat the ionophoretically induced glutamate

0 2.6 3.0 3.4 3.8 4.2 4.6Stimulus Intensity (volts)

Figure 2. Kynurenate educes 2 in a dose-dependentannernde-pendent f stimulusntensity,whileNl isunaffected. he graphs howthe amplitudes f Nl (A) and N2 (B) asa function of stimulationn-tensity rom anexperimentn which4 differentconcentrationsf kyn-urenate 0.2, 0.4, 0.8, and 1.6 mM) were appliedsuccessivelyo thesame lice.Note that after the washout f 1.6mM kynurenate, he N2amplitudeeturned o its controlvalues.

It is generally accepted hat the major glutaminergic nput tothe neostriatum derives from cortical afferents (Dray, 1980;Graybiel and Ragsdale, 1983). Therefore, it is likely, as sug-gested reviously (Cordingley and Weight, 1986), hat N2 is dueto the activation of cortical afferentsand the subsequent eleaseof glutamate. Consistent with this hypothesis was our findingthat we could elicit a kynurenate-sensitive N2 component withthe stimulating microelectrode placeddirectly in the corpuscal-losum.DA has been reported to decrease he releaseof glutamatefrom neostriatal slices Mitchell and Doggett, 1980; Rowlandsand Roberts, 1980)and receptor binding studieshave suggestedthat the terminals of neocortical afferentspossess A receptors(Schwartz et al., 1978).However, we found that bath applicationof DA [l-100 PM with (n = 8) or without (n = 7) 0.3 mM


4/7

The Journal of Neuroscience, October 1988, &IO ) 3753

A CONTROL

D ATROPINE

kEl CARBACHOL c \IVASH

F KYNURENATEATRO+CARB

field potential is unaffected by carbachol at a time when thestimulus-evoked N2 component is significantly reduced. Kyn-urenate, unlike carbachol, reduced both field potentials simul-taneously with a similar time course, demonstrating that thefield potential recorded in response to ionophoretic applicationof glutamate was indeed due to the activation of functionalglutamate receptors. These results are consistent with the hy-pothesis that muscarinic receptors localized on the terminals ofcorticostriatal afferents can inh ibi t transmitter release.Adenosine has been found to presynaptically inh ibi t excitato-ry synaptic transmission in a variety of brain regions (Okadaand Ozawa, 1980; Kocsis et al., 1984; Dunwiddie and Haas,1985). Both in situ radioligand binding studies (Goodman andSnyder, 1982) and electrophysiological studies (Kostopoulos andPhill is, 1977; Trussell and Jackson, 1985) suggest that there aremoderately high concentrations of adenosine receptors in theneostriatum. Furthermore, adenosine antagonists have loco-motor stimulatory effects, suggesting a role for adenosine in thecontrol of motor activity (Daly et al., 198 1). Therefore, we ex-amined the actions of adenosine on N 1 and N2. Like kynurenateand carbachol, adenosine (0.1-0.5 mM) reversibly decreased orabolished N2 without significantly affecting Nl (Fig. 6, A-C, n= 8). Appl ication of theophyl line (0.1-0.3 mM), a well-estab-lished adenosine antagonist (Daly et al., 1981) blocked the re-duction of N2 by adenosine (Fig. 6E; n = 3), demonstrating thatthis action is mediated by adenosine receptors.To determine whether adenosine was acting primarily pre- orpostsynaptically to reduce N2, we again util ized the procedureof recording simultaneously the synaptically evoked N2 and thefield response to ionophoretically applied glutamate. Figure 7Bshows that at a time when adenosine had significantly reducedN2, the glutamate field potential was unaltered (n = 3). In con-trast, kynurenate reduced both field potentials simultaneously(Fig. 70). Thus, like carbachol, adenosine likely reduces or blocksN2 by acting presynaptically to inh ibi t transmitter release.

Figure 4. Muscarinic receptor acti-vation reduces N2. A-C, Application ofcarbachol(l0 PM ) reversibly reducedN2without affecting Nl. D, E, Subsequentapplication of atropine (2 PM ) had noeffect on the field by itself, but blockedthe action of carbachol. F, After wash-out ofcarbachol and atropine, kynuren-ate (1 mM) blocked N2.

DiscussionThe corticostriatal projection comprises the major input to theneostriatum using glutamate as the neurotransmitter (Dray, 1980;Graybiel and Ragsdale, 1983). In agreement with previous sug-gestions (Takagi and Yamamoto, 1978; Cordingley and Weight,1986) we have presented evidence that the second negativity(N2) of the locally evoked field potential in neostriatal slicescorresponds primari ly to the synaptic activation of striatal neu-rons by cortical afferents. Thus, 3 different excitatory aminoacid antagonists (kynurenate, piperidine dicarboxylate, and Y-D-glutamylglycine) reversibly reduce or block N2. Furthermore,direct stimulation of the corpus callosum also elicits a kyn-urenate-sensitive N2 component of the field potential.Anatomical studies have demonstrated that the only neo-striatal afferents that run through the corpus callosum comefrom neocortex (Carman et al., 1965; Carpenter and Sutin, 1983).Thalamic afferents pass directly into the neostriatum (Powelland Cowan, 1956; Carpenter and Sutin, 1983) and would notbe activated by callosal stimulation. There is no evidence tosupport the existence of intrinsic glutaminergic neurons or af-ferent sources of excitatory amino acids other than cerebralcortex and thalamus. Afferent or intrinsic pathways not util izingexcitatory amino acid transmitters are also likely stimulatedwhen the cortical afferents are activated, but the virtual absoluteblockade of N2 by kynurenate strongly suggests that N2 over-whelmingly reflects the synaptic responses of neostriatal neuronsto the activation of cortical afferents.Misgeld and colleagues (1980) have previously presented evi-dence that nicotinic antagonists can reduce N2. However usingthe same nicotinic antagonists as Misgeld et al. (1980), we foundno effect on N2. We have no explanation for this discrepancy,except to note that neither Cordingley and Weight (1986) norTakagi and Yamamoto (1978) found that nicotinic antagonistshad any significant actions on N2.


5/7

3754 Malenka and Kocsis l Presynaptic Inhibition of Corticostr iatal Pathwa y

A CONTROL B CARBACHOL I 1 I

* GLUTAMATEv vWASH I-vv

Figure 6. Adenosineeceptor ctiva-tion reduces 2. A-C, Applicationofadenosine0.2mM) reversibly educedN2 without alteringNl. D, E, The-ophylline 0.1 mM)by itselfhadno ef-fect on he ieldbut nhibited heactionof adenosine,emonstratinghat aden-osine as cting t adenosineeceptors.F, N2 wassubsequentlylocked y pi-peridine icarboxylatePDA; 1 OmM),an excitatory amino acid antagonist,againdemonstratinghat N2 is me-diatedby a glutamate-like ompound.

D KYNURENATE1

2

* CONTROL

Muscarinic agonistshave consistently been found to reduceN2 (Takagi and Yamamoto, 1978; Weiler et al., 1984), as wellas ntracellularly recordedEPSPs Dodt and Misgeld, 1986) butalso have significant direct depolarizing actions on striatal neu-rons (Dodt and Misgeld, 1986). Using the technique of simul-taneously recording the synaptically evoked N2 and the fieldpotential evoked by the ionophoretic application of glutamate,we found that while both responses ere reducedby kynurenate,only N2 was affected by carbachol. We interpret this result asindicating that carbachol s acting presynaptically to inhibit cor-ticostriatal synaptic transmission.Further support for this hy-pothesiscomes rom the observation that carbachol,at the dosesused n this study, hasno effect on either the membranepotentialor input resistance f striatal neurons Dodt and Misgeld, 1986).ACh, acting on muscarinic receptors, has been shown to in-hibit synaptic transmissionat synapses tilizing excitatory ami-no acids n a variety of systems, ncluding rat ventral horn cells(Jiang and Dun, 1986) and hippocampus (Hounsgaard, 1978;Valentino and Dingledine, 1981). The presynaptic inhibitoryaction of ACh may be due to either a reduction in calciumconductance or to an increase n potassium conductance oc-curring at or near the axon terminal (North, 1986).The validity of our conclusion concerning the locus of car-tFigure 5. Carbachol ctspresynapticallyo reduceN2. The field po-tential in responseo electricalstimulation I) and the responseoionophoretic pplication f glutamate 2) recordedrom the sameo-cation areshown ollowingcarbachol nd kynurenate pplication. -C, Carbachol10 M) reversibly educedN2 without affecting he glu-tamate esponse, hileapplication f kynurenate0) to thesameieldsreduced oth simultaneously.

D THEOPHYLLI.NE

B ADENOSINE

1.5 mV9 msec WASH

E ADEN+THEOPH PDA


6/7

The Journal of Neuroscience. October 1%38, 8(lQ) 3755

A CONTROL B ADENOSINE

* GLUTAMATE

C WASH

IT+----2

D KYNUAENATE1 I-

2

Figure 7. Adenosine cts resynapticallyo inhibit corticostriatal yn-aptic transmission.-C, Adenosineeversibly educedN2 (I), whilethe simultaneouslyecordedieldpotential n responseo ionophoreticapplication f glutamate2) wasunaffected. , In contrast,kynurenate(1mM),appliedo thesame lice, educed othN2 and he ieldpotentialin responseo glutamate pplication.bachols action is dependent on the assumption hat the samegroup of postsynaptic glutamate receptors s being activated bysynaptically released ransmitter and the exogenously appliedglutamate. This seemsikely, sinceboth responses ere ecordedvia the same ecording electrode, but it is conceivable that the2 responses eflect the activation of different populations ofreceptors,with different sensitivities to the postsynaptic actionsof carbachol.

The biochemical anatomy of the neostriatum has received agreatdeal of attention (Graybiel and Ragsdale,1983).One fairlywell-developed classification, based primarily on staining forAChE, divides the neostriatum into striosomesand matrix.Striosomesare areas hat stain poorly for AChE (Graybiel andRagsdale,1978) exhibit densepatchesof opiate receptor bind-ing (Herkenham and Pert, 198 ), and are enriched in immu-noreactivity for enkephalin and substanceP (Graybiel et ai.,1981). The extrastriosomal matrix exhibits high AChE histo-chemical staining (Graybiel and Ragsdale,1978) and high im-munoreactivity to choline acetyltrasferaseGraybiel et al., 1986)suggestinghat the cholinergic neuropil is concentrated primar-ily in the matrix. In our experiments, carbachol always educedN2 independent of microelectrode placement. This may reflect(1) the fact that the striosomal-matrix distinction becomesmuch

lessdramatic in adult animals (Graybiel and Ragsdale,1983),(2) the inability of our recording techniques to discern subtlespatial differences n the density of muscarinic eceptorspresentin neostriatum, or (3) our selection criteria for good fieldpotentials inadvertently results n the consistent placement ofthe recording electrode n the matrix.Like carbachol, adenosine educed or blocked N2 at a timewhen the glutamate field responsewas unaffected, suggestingthat adenosinemay also presynaptically inhibit synaptic trans-mission betweencortical afferents and neostriatum. Adenosinehas a similar action in the hippocampus, where it has beenshown to inhibit the release f glutamate (Dolphin and Archer,1983; Corradetti et al., 1984) and to presynaptically inhibitexcitatory synaptic transmission (Okada and Ozawa, 1980;Dunwiddie and Haas, 1985). Functional adenosine eceptorsare also found on the terminals of the cerebellar parallel fibers(Kocsiset al., 1984). n cultured neostriatal cells,adenosine asbeen shown to increase membrane conductance to potassium(Trussell and Jackson, 1985). If adenosinehad the sameactionon presynaptic nerve terminals, it would be expected o decreasestimulus-evoked neurotransmitter release.However, adenosinemay also have actions on the voltage-dependent calcium con-ductance at the terminal (Wakade and Wakade, 1978) or inter-fere intracellularly with the calcium-stimulated secretory pro-cess Silinsky, 1984).All areasof the cerebral cortex project to the striatum topo-graphically. However, the corticostriatal projection is not uni-form, but derives, rather, from a number of distinct subclassesof cortical neuronswith different axonal collateralsand projec-tions (Joneset al., 1977; Wilson, 1987). Furthermore, diversecortical regionsmay project to the samestriatal zones Selemonand Goldman-Rakic, 1985; Malach and Graybiel, 1986). Thesegregation f receptors or ACh or adenosine n distinct subsetsof corticostriatal fibers would permit the selective nhibition ofinput to the striatum from diverse regionsof cerebral cortex orfrom distinct classes f cortical neurons. Quantitative autora-diography of binding sites following selective cortical lesionswill help determine whether all or a subset of corticostriatalfibers possesseceptors or ACh or adenosine.ReferencesCarman, . B., W. M. Cowan, . P. S. Powell, ndK. E. Webster 1965)A bilateral ortico-striate rojection. . Neurol.Neurosurg. sychiatry28: 7 -77.Carpenter,M. B., andJ. Sutin (1983)Human Neuroanatomy, 8th Ed.,vv. 587-589.Williams& Wilkins. Baltimore.MD.Cdrdingley,G.E., andF. F. Weight (1986) Non-choline& synapticexcitation n neostriatum: harmacologicalvidence or mediation

by a glutamate-likeransmitter.Br. J. Pharxnacol.8: 847-856.Corradetti,R., G. Moneti, F. Moroni, G. Pepeu,and A. Wieraszko(1984) Adenosine ecreasesspartatendglutamateeleaserom rathippocampallices.Eur. J. Pharmacol. 04: 19-26.Daly, J. W., R. F. Bruns, ndS.H. Snyder (1981) Adenosineeceptorsin the centralnervous ystem:Relationshipo the centralactions fmethylxanthines.ife Sci. 28: 2083-2097.Dodt, H. U., andU. Misgeld (1986) Muscarinicslowexcitationandmuscarinicnhibitionof synaptic ransmissionn the at neostriatum.J. Physiol. Lond.)380:593-608.Dolphin,A. C., andE. R. Archer (1983) An adenosinegonistnhibitsanda cyclic AMP analoguenhanceshe releasef glutamate ut notGABA from slices f rat dentate yms.Neurosci. ett. 43: 49-54.Dray, A. (1980) The physiology nd pharmacology f mammalianbasal anglia.Prog.Neurobiol. 4: 221-335.Dunwiddie,T. V., andH. L. Haas (1985)Adenosinencreasesynapticfacilitation n the n vitro rat hippocampus:videnceor apresynapticsiteof action.J. Physiol. Lond.)369:365-377.


7/7

3756 Malenka and Kocsis * Presynaptic Inhibition of Corticostr iatal Pathwa y

Ganong, A. H., T. H. Lanthom, and C. W. Cotman (1983) Kynurenic Malach, R., and A. M. Graybiel (1986) Mosaic architecture of theacid inhibits synaptic and acidic amino acid-induced responses in the somatic sensory-recipient sector of the cats striatum. J. Neurosci. 6:rat hippocampus and spinal cord. Brain Res. 573: 17&l 74.Ginsborg, B. L., and G. D. S. Hirst (I 972) The effect of adenosine onthe release of the transmitter from the phrenic nerve of the rat. J.Physiol. (Lond.) 224: 629-645.Goodman. R. F.. and S. H. Snvder (1982) Autoradioeranhic localiza-

3436-3458.Misgeld, U., Y. Okada, and R. Hassler (1979) Locally evoked poten-tials in slices ofrat neostriatum: A too! for the investigation ofintrinsicexcitatory processes. Exp. Brain Res. 34: 575-590.Misaeld, U., M, H, Weiler. and I. J. Bak (1980) Intrinsic choline&ction o f adenosine receptors id rat brain~using [Hlcyclohexyladenosine.J. Neurosci. 4: 1230-1241.Graybiel, A. M., and C. W. Ragsdale, Jr. (1978) Histochemical ly dis-tinct compartments in the striatum of human, monkey, and cat dem-onstratedby acetycholinesterase staining. Proc. Natl. &ad. Sci. USA75: 5723-5726.

e&tation in the rat ncostriatum: Nicotinic and muscarinic receptors.Exp. Brain Res. 39: 401-409.Misgeld, U., M. H. Weiler, and D. K. Cheong (1982) Atropine en-hances nicotinic cholinergic EPSPs in rat neostriatal slices. Brain Res.253: 3 17-320.Mitchell, P. R., and N. S. Doggett (1980) Modulation of striatal [HI-glutamic acid release by dopaminergic drugs. Life Sci. 26: 2073-208 1.North, R. A. (1986) Muscarinic receptors and membrane ion con-ductances. Trends Pharmacol. Sci. (Suppl.) 7: 19-22.Okada, Y., and S. Ozawa (1980) Inhibitory action of adenosine onsynaptic transmission in the hippocampus of the guinea pig in vitro .Eur. J. Pharmacol. 68: 483492.

Graybiel, A. M., and C. W. Ragsdale, Jr. (1983) Biochemical anatomyof the striatum. In Chemical Neuroanatomy, P. C. Emson, ed., pp.427-504, Raven, New York.Graybiel, A. M., C. W. Ragsdale, Jr., E. S. Yoneoka, and R. P. Elde(198 1) An immunohistochemical study of enkephalins and otherneuropeptides in the striatum of the cat with evidence that the opiatepeptides are arranged to form mosaic patterns in register with thestriosomal compartments visible by acetylcholinesterase staining.Neuroscience 6: 377-397.Graybiel, A. M., R. W. Baughman, and F. Eckenstein (1986) Cholin-ergic neuropil ofthe striatum observes striosomal boundaries. Nature323: 625-627.Herkenham, M., and C. B. Pert (198 1) Mosaic distribution of opiatereceptors, parafascicular projections and acetylcholinesterase in ratstriatum. Nature 291: 415418.Hounsgaard, J. (1978) Presynaptic inhibitory action of acetycholinein area CA1 or the hippocampus. Exp. Neurol. 62: 787-797.Jahr, C. E., and K. Yoshioka (1986) Ia afferent excitation of moto-neurones in the in vitro new-born rat spinal cord is select ively antag-onized by kynurenate. J. Physiol. (Lond.) 370: 5 15-530.Jiang, Z. G., and N. J. Dun (1986) Presynaptic suppression of excit-atory postsynapt ic potentials in rat ventral horn neurons by muscar-inic agonists. Brain Res. 381: 182-186.Jones, E. G., J. D. Coulter, H. Burton, and R. Porter (1977) Cells oforigin and terminal distribution of corticostriatal fibers arising in thesensory-motor cortex of monkeys. J. Comp. Neural. 173: 53-80.Joyce, J. N., and J. F. Marshall (1987) Quantitative autoradiographyofdopamine D, sites in rat caudate-putamen: Localization to intrinsicneurons and not to neocortical afferents . Neuroscience 20: 773-795.

Powell, T. P. S., and W. M. Cowan (1956) A study ofthalamo-striaterelations in the monkey. Brain 79: 364-390.Rowlands, G. J., and P. J. Roberts (1980) Activation of dopaminergicreceptors inhibits calcium-dependent glutamate release from cortico-striatal terminals in vitro . Eur. J. Pharmacol. 62: 241-242.Schwartz, R., I. Creese, J. T. Coyle, and S. H. Snyder (1978) Dopaminereceptors localized on cerebral cortical afferents to rat corpus striatum.Nature 271: 766-768.Selemon, L. D., and P. S. Goldman-Rakic (1985) Longitudinal to-pography and interdigitation of corticostriatal projections in the rhe-sus monkey. J. Neurosci. 5: 776-794.Silinsky, E. M. (1984) On the mechanism by which adenosine receptoractivation inhibits the release of acetylcholine from motor nerve end-ings. J. Physiol. (Lond.) 346: 243-256.Takagi, M., and C. Yamamoto (1978) Suppressing action of cholin-ergic agents on synaptic transmissions in the corpus striatum of rats.Exp. Neurol. 62: 433443.Trugman, J. M., W. A. Geary, II, and G. F. Wooten (1986) Localiza-tion of D-2 dopamine receptors to intrinsic striatal neurones by quan-titative autoradiography. Nature 323: 267-269.Trussell, L. O., and M. B. Jackson (1985) Adenosine-activated po-tassium conductance in cultured striatal neurons. Proc. Natl. Acad.Sci. USA 82: 48574861.Kemp, J. M., and T. P. S. Powell (1971) The site of termination ofafferent fibers in the caudate nucleus. Phil. Trans. R. Sot. Lond. [Biol.]262: 4 13-427.

Valentino, R. J., and R. Dingledine (198 1) Presynaptic inhibitory eff ectof acetylcholine in the hippocampus. J. Neurosci. I: 784-792.Wakade. A. R.. and T. D. Wakade (1978) Inhibition of noradrenalineKita, H ., T. Kita, and S. T. Kitai (1985) Active membrane propertiesof rat neostriatal neurons in an in vitro slice preparation. Exp. BrainRes. 60: 54-62.Kocsis, J. D., D. L. Eng, and R. B. Bhisitkul (1984) Adenosine selec-tively blocks parallel-fiber-mediated synaptic potentials in rat cere-bellar cortex. Proc. Natl. Acad. Sci. USA 81: 6531-6534.Komhuber, J., and M. E. Komhuber (1986) Presynaptic dopaminergicmodulation of cortical input to the striatum. Life Sci. 39: 669-674.Kostopoulos, G. K., and J. W. Phillis (1977) Purinergic depression o fneurons in different areas of the rat brain. Exp. Neural. 55: 7 19-724.

release by adenosine. J. Physiol. (Land:) 282: 3549.Weiler M. H., U. Misgeld, and D. K. Cheong (1984) Presynapticmuscarinic modulation of nicotinic excitation in the neostriatum.Brain Res. 296: 11 l-120.Wilson, C. J. (1987) Morphology and synaptic connections of crossedcorticostriatal neurons in the rat. J. Comp. Neurol. 263: 567-580.Yamamoto, C. (1973) Neurotransmitteractions studied in brain slices.Adv. Neurol. Sci. 17: 64-70.

robert c. malenka and jeffery d. kocsis- presynaptic actions of carbachol and adenosine on...

Documents