Brain Research, 586 (1992) 195-202 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

195

BRES 17922

Rapid changes in ascorbate and dopamine release in rat nucleus accumbens after intracerebroventricular administration of NMDA L. Svensson, C. W u , P. H u l t h e , K. J o h a n n e s s e n a n d J.A. Engel Department of Pharmacology, Universityof Gi~teborg,Giiteborg (Sweden) (Accepted 25 February 1992)

Key words: Dopamine; Ascorbate; N-MethyI-D-aspartic acid (NMDA); Intracerebroventricular administration; Nucleus accumbens; In vivo voitammetu; Spreading depression

In vivo voltammetry at electrochemically pretreated carbon fibre electrodes was used to investigate the effect of intracerebroventricular (i.c.v.) administration of N-methyI-D-aspartic acid (NMDA) on neuronal activity in rat nucleus accumbens. Infusion of a low dose of NMDA (1 nmol) was followed a few minutes later by rapid changes in both Peak 1 and Peak 2 heights indicating large but short-lived increases in the extracellular concentrations of ascorbate and catecholamines, respectively. These responses did not seem to be dependent on the dose infused since infusion of NMDA for a longer time period neither changed the amplitude nor the time-course of these effects. The increase in Peak 2 height was resistant to pargyline pretreatment indicating that this response mainly reflected the release of dopamine. The administration of NMDA was followed by behavioural activation in the animals but not convulsions. Co-administration of the competitive NMDA receptor antagonist, CPP (1 nmoi), completely blocked these effects while the acetylcholine receptor antagonist, atropine (1.5 nmol), and the GABA receptor antagonist, picrotoxin (1 nmol), failed in this respect. The phenomenon spreading depression is discussed as a possible explanation of these results.

INTRODUCTION The nucleus accumbens is considered to constitute a functional interface between the limbic system and the motor system 2s. As such much interest has been focused on the possible involvement of this structure in e.g. locomotor activity, positive reinforcement and psychiatric disorders ts,a6,3~. In support of these assumptions, local infusion of the glutamate analogue, NmethyI-D.aspartic acid (NMDA), or dopamine (DA), produces locomotor stimulation in the rat 2,33. Con= versely, these effects can be blocked by DA receptor antagonists. Furthermore, local infusion of DA receptor antagonists blocks self-administration of dependence producing drugs like cocaine and amphetamine 34 and electrical self-stimulation 3s. A large body of evidence indicates that L-glutamate (GLU) acts as an excitatory neurotransmitter in the

central nervous system. The nucleus accumbens receives ascending glutamatergic projections mainly from the cortical areas the prefrontal cortex, hippocampus and amygdala and the subcortical thalamic nucleus %~s. This area also receives ascending dopaminergic projections mainly originating in the ventral tegmental area (area A10 according to Dahlstr0m and Fuxe "J). Thus, the nucleus accumbens is a possible site for interactions between these two neurotransmitter systems. A reciprocal influence between GLU a~ld DA in this area is also indicated in a number of studies to,a6. The functional organization of nucleus accumbens is still obscure, however, and the way glutamatergic and dopaminergic mechanisms interact in this area is much debated. In the present study the effects of intracerebroventricular (i.c.v.) administration of NMDA on AA and DA release in nucleus accumbens was investigated using in vivo voltammetry in freely moving rats. The

Correspondence: L. Svensson, Department of Pharmacology, University of G6teborg, P.O.B. 33031, S-400 33 G6teborg, Sweden. Fax: (46) (31) 821795.

196 A (

H .

I

E-

--,J

Fig. I. Schematic description of the electrode imphmt:ltion assembly used in the present experiments. A. working electrode; B. reference electrode; C. auxilia~' electrode: D. copper wire; E. screw; F. threaded cap; G. bra,~ rod; !i. plastic tube; I. dental cement: J. skull; K, plastic pedestal; L. guide cannula.

interaction between GLU and non-GLU mechanisms in this area was investigated by co-administration of selective antagonists. MATERIALS AND METIIODS

ha t'it'o t'oltammetn' The animals used were malt: Sprague-Dawley rats (ALAB, Sollentuna, Swede:,1, 4(1(I-4511g). The rats were anaesthetized with a mixture of Ketalar (9(1 mg/kg, i.p.) and Rompun (Ill mg/kg, i.p.) and placed in ,'1 stereotaxic frame according to the recommendations o1' Paxinos and Watson .a~, The body temperature wits controlled by a heating pad during surgery. A schematic description of the electrode implantation assembly is shown in Fig. 1. The skull (J) was exposed and two screws (El were fastened. An insulated silver wire was attached to one of the screws to serve as auxiliary electrode (C). A hole was drilled in the skull and an Ag/AgCI reference electrode (B) was lowered into the hole to the level just below the dura mater. A second hole was drilled over nucleus accumbens and the dura mater was carefully cut away, A guiding device for insertion and retraction of the working electrode was positioned over the hole. This was locally constructed and consisted of two parts, A plastic tube (H) with threads cut on the outside and 4 longitudinal tracks cut on the inside, A second part consisted of a brass rod (G) with its middle part shaped as a square bar. The 4 edges of this bar fitted to the longitudinal tracks inside the plastic tube, The rod could then be moved inside the tube without revolving by means of a threaded cap (F), This was freely attached to the distal end of the rod by an o-riug, The

working electrode (A) was mounted inside a cavity in the rod and could be advanced into the brain by revolving the cap. A third hole was drilled over the ipsilateral ventricle and a stainless steel cannula serving as a guide cannula (L) for the i.c.v, infusion was positioned over the hole. In order to prepare for later electrical connection to the recording equipment the auxiliary and reference electrodes were connected to gold-plated socket contacts on a plastic pedestal (K) (MS 363, Plastic One, Roanoke, VA, USA). A short insulated copper wire (D) was connected to a third socket contact and the opposite end of the wire bent away from the skull and left disconnected. At the end of surgery the electrode guiding device, the stainless steel cannula anti the pedestal were fixed to the skull by dental cement (I) and the animal left to recover for 2-3 days. A cylindrical carbon fibre electrode prepared from a 30 #m diameter carbon fibre (TEXTRON Specialty Materials, Lowell, MA, USA) was used as working electrode. The carbon fibre was sealed with epoxy resin in a glass capillary and cut to a protruding length of 0.5 ram. The electrode was electrochemically treated in phosphate buffered saline (PBS) before implantation according to a slightly modified technique by Gonon and co-workers ~7. A triangular wave potential of 0 to -3,0 V, 70 Hz was applied for 5 s followed by 0 to + 3.0 V, 7(1 Hz for 20 s finally followed by a constant potential of + I.SV DC Ibr 2(} s, Calibration recordings in vitro revealed a first current peak (Peak 1)at -0.1 V corresponding to the oxidation of AA (211(I uM) and a second peak (Peak 2) at +ILl V corresponding to either the oxidation of DA (I ~M) or DOPAC (10 ~M), The shape of these peaks were only slightly modified during in vivo recordings, At the day of recording the working electrode was slowly moved into the brain of the fully awake animal using the guiding device, The coordinates used for nucleus accumbens were: + 2,0 mm anterior to bregma (A), + 1,2 mm lateral to median line (L) and - 6 , 8 mm below brain surface (V), After implantation the electrode wire was soldered to the copper wire previously fixed to the plastic pedestal and the pedestal connected to an armed 6-lead cable (6TCS, Plastic One, Roanoke, VA, USA) which in turn was connected to the recording equipment via a rotational swivel. The animal was thereafter immediately placed in a circular arena (diameter 40 cm) and in vivo recordings initiated. The voitammograms were produced by square wave voltammetry 3, using a microcomputer controlled data acquisition unit (GVI-4CH Voltammetric Instrument, Metod och Produkt Svenska AB, G6teborg, Sweden). The specific qualities of this technique are further

197 presented in a previous paper l l. The recording parameters were as follows: Potential sweep - 2 0 0 mV to + 200 mV, sweep rate 125 mV/s, square wave pulse modulation 50 mV, step potential 5 mV, pulse frequency 25 Hz and scan repetition period 60 or 20 s. The working electrode was physically disconnected between scans. The current was sampled both at the end of the forward and the reverse pulse of the square wave. A true differential current was obtained by subtracting the latter current from the former. The peak heights were stable within 30 rain of the initiation of in vivo recordings and this period was allowed to pass before experimental treatment was begun. Intracerebroventricular infusion was performed by inserting an infusion cannula (outer diameter 250/zm) through the previously fixed guide cannula. The coordinates used for the lateral ventricle were: 0.0 mm anterior to bregma (A), + 1.2 mm lateral to midline (L) and - 4 . 0 mm below brain surface (V). Infusion was immediately initiated at an infusion rate of 0.5/~i/min by use of a microinfusion pump (CMA/100, Carnegie Medicine AB, Stockholm, Sweden). The total volume infused was always 1 /zl. After termination of the infusion the cannula was allowed to remain on site for l rain before retraction.

Drugs N-MethyI.D-aspartic acid (NMDA) (Sigma Chemical Co, USA), 3-((RS)-2-carboxy-piperazin.4-yl)-propyl-lphosphonic acid (CPP) (Tocris Neuramin, UK), pargyline hydrochloride (Sigma Chemical .Co, USA), at-

70

~ 8

2 1

60 50-

40: 30-

1

"~ .'g 20. 10

2

NMDA Response ..... -

-200

Baseline

i

-100 0 Applied Potential (mV)

100

Fig. 2. A recording showing the effects of an i.c.v, infusion of NMDA (1 nmol) on the current peaks (Peak 1 representing the oxidation of AA and Peak 2 representing the oxidation of catecholamines) measured by in vivo voltammetry in rat nucleus accumbens. The scan showing the maximum current change is shown. The scan just prior to the initiation of the infusion (baseline recording) is included for comparison.

350=~ 300.F_.

-....

n• t,-

150

Peak 2

/\

25o-

200

Peak 1

::

/X /..."...

,ool_.__..

,'gx 50] O

NMDA .... I

0

~

'

I

60

'

'

I

'

'

I

120 180 Time (sec)

J

'

I

240

Fig. 3. The time-course of the changes in Peak 1 and Peak 2 heights shown in Fig, 2. The time period for the infusion of NMDA is indicated by the bar in the figure. Scan repetition period was 20 s.

ropine sulphate (ACO L~ikemedel, Sweden) and picrotoxin (Sigma Chemical Co, USA) were used. All drugs were dissolved in 0.9% saline.

RESULTS Intracerebroventricular infusion in awake, freely moving rats of NMDA (0.5/zl/min) at a total dose of 1 nmol produced rapid changes in both Peak 1 and Peak 2 heights as measured by in vivo voltammetry in nucleus accumbens (Figs. 2 and 3). Large increases in currents of short durations were recorded indicating pronounced but short-lasting increases in the release of AA and catecholamines, respectively. These responses were accompanied by a 30 to 45 mV negative shift in oxidation potentials (Fig. 2). The oxidation potential is influenced by the concentration of the species being oxidized, the pH and local variations in the ionic composition of the extracellular fluid. According to the Nernstian equation, the 30 to 45 mV negative shift in oxidation potentials can be explained by a 10 to 40 times increase in the concentrations of AA and catecholamines. Two (Fig. 4) or 3 (data not shown) successive NMDA infusions at 45 rain intervals evoked very similar responses with no signs of response attenuation but successive infusions at 5 min intervals failed to evoke new responses. Responses were also frequently observed after NMDA infusions at a total dose of 0.5 nmol (data not shown). Between the infusions the currents rapidly returned to pre-infusion values. Prolonged infusions over a time period longer than 2 min did not change the time-course of these effects but eventually induced convulsions in the rats.

198 Baselinecurrent NMDAevoked Baselinecurrent NMDAevoked beforestart of peak current afterpargyline peak current (no pargyline) treatment after pargyline (n = 5) treatment(nA) Peak

300

o~ ¢., O~

25O

I-'1 1st Infusion

Peak 2

[~J 2nd Infusion

"~ 200 -1,,e

13.23-1-2.42 ¢¢'A' 0.47+0.20

9.46+ 1.52'k~' (n= 1)

v e-

m 150

T

8-

6C.3 fq 4-

a.

co 100 .'g x O

6.40+ ! . 5 4 2-

(D a.

50

2- NMDA 0

l

0

I

10

Pargyline i

I

20

0 Peak 1

'

I

30

'

I

,

I

l

I

40 50 60 Time (rain)

l

I

l

70

I

80

,

I

90

Peak 2

Fig. 4. The effects of 2 i.c.v, infusions of NMDA (! nmol) administered with a 45 min interval to the same group of rats (n = 5) on the current peaks measured by in vivo voltammetry in nucleus accumhens. Shown are the means + S.E.M.

in a number of experiments different drugs were administered either i.c.v, or intraperitoneally (i.p.) in order to investigate their influence on the NMDAevoked current responses, in a first series of experiments NMDA was infused i.c.v. 2 times to 5 rats and pargyline was administered i.p. in the intermediate time period 15 min after the first NMDA infusion and 45 rain before the second infusion. Pargyline treatment caused a progressive decrease in Peak 2 height but no effect on Peak l height (data not shown), Infusion of NMDA caused similar current responses at both occasions. A representative current versus time plot on the effects of these treatments on Peak 2 is shown in Fig. 5 together with a full statistical analysis. in subsequent experiments the rats first received a single i.c,v, NMDA (1 nmol) infusion to verify the occurrence of responses. After a time period of 30-45 rain when the currents had returned to normal the baseline peak heights were measured and the rats received an i.c.v, infusion of saline, CPP (1 nmol), atropine (1.5 nmol) or picrotoxin (l nmol). Ten rain

Fig. 5. Representative current versus time plot showing the effects of an i.c.v, infusion of N M D A (1 nmol) on the Peak 2 current before and after pargyline (75 mg/kg) treatment measured by in rive voltammetry in rat nucleus accumbens (lower panel). NMDA was infused i.c.v. 2 times and pargyline was administered i.p. in the intermediate time period 15 rain after the first N M D A infusion and 45 rain before the second infusion. The points of time for administration of the respective drugs are indicated by the arrows in the figure. Scan repetition period was 60 s. A full statistical analysis of these effects in 5 rats is shown in the upper panel. ** P < 0,01 compared to the baseline current before start of the N M D A infusion. Two-tailed Student's t-test.

later the peak heights were measured again and the rats received a second infusion of NMDA (1 nmol). The maximum changes in peak heights recorded during the following 3 rain period were assessed and compared with the respective tare-infusion values. CPP was found to completely block the NMDA-induced increases in both peak heights while atropine and picrotoxin did not significantly change any of these responses, These results are presented in Table 1. Gross observations indicated that the i.c.v, infusion of NMDA and picrotoxin caused behavioural activation in the animals. This consisted mainly of well coordinated locomotor stimulation. In the case of NMDA, also jumping, sniffing and grooming behaviour were sometimes observed. Convulsions were not ob-

TABLE I

The effect of carious drugs administered i.c.c, on the current peaks measured by in r'i~'o ~'oltammetry in rat nucleus accumbens When combined with NMDA infusion, saline, CPP, atropine and picroto×in were administered i.c.v. 10 rain before NMDA. The peak heights recorded just prior to the infusion of N M D A were used to determine the effects of the pretreatment alone. These values were also used as pro-drug values to determine the effects of the NMDA infusion.

Dr~,g refused

Clzange m Peak I heiRh¢ (%)

Change in Peak 2 height (%)

n

n,r,

n,r, 181 +34 ** !1 + 10

5 5 5

n,r, n.r,

5 4

157 + 26 * * 36+ 9 * 198 + 20 * *

4 4 4

.....

Saline NMDA (1 nmol) after saline infusion CPP (I nmol) NMDA after CPP infusion Atropine (!.5 nmol) NMDA after atropine infusion Picrotoxin (I nmol) NMDA after picrotoxin infusion

56:1:16 * n,r.

n,r, n,r, 45 + 13 * -20+ 6 * 38 :t: 13

n.r, = no response, * P < 0,05, * * P < 0,01 compared to the pre-drug peak height, Two-tailed Student's t-test,

199 served at the doses used. Intracerebroventricular administration of CPP, on the other hand, seemed to cause a slight sedation. No attempts to quantify these effects were made. DISCUSSION The in vivo voltammetry recordings performed in this study demonstrate that i.c.v, administration of a low dose of NMDA produces transient increases in the release of AA and catecholamines in the nucleus accumbens of awake and freely moving rats. As to the nature of the detection of catecholamines it is important to note that the working electrode used with the present method does not normally differentiate between DA and its main metabolite, DOPAC. Nevertheless, previous studies have shown that in nucleus accumbens Peak 2 reflects the extracellular concentration of DOPAC with little influence from DA or other species 4. The electrochemically modified carbon fibre electrode is more sensitive to DA but this property is overshadowed by a much higher basal extraceilular concentration of DOPAC. In order to identify the electroactive species responsible for the NMDA-evoked change in Peak 2 height recorded in the present study a series of experiments was performed in rats where the formation of DOPAC was blocked by the monoamine oxidase inhibitor, pargyline. This treatment markedly reduced Peak 2 height verifying that this peak mainly represents the oxidation of DOPAC under basal conditions. However, i.c.v, infusion of NMDA still produced a large increase in Peak 2 height in these animals. In fact, no reduction in the current response to NMDA was observed in parbvline-treated animals as compared to the response in saline-treated animals (NMDA-evoked response in pargyline-treated animals: 8.99 + 1.42 nA, response in saline-treated animals: 6.83 :l: 1.07 nA). Thus it can be concluded that the increase in Peak 2 height recorded mainly represented the oxidation of DA. In vitro calibration of the working electrode in known concentrations of DA indicated that NMDA produced at least a 20-fold increase in extracellular DA concentration. In other experiments we have found that similar responses can be recorded also in rat striatum. Friedmann and Gerhardt ~3, using in vivo voltammetry, have also shown that local application of NMDA in the striatum evokes DA release. However, previous experiments using in vivo dialysis technique have failed to show changes in the release of DA in this area after i.c.v, infusion of NMDA 2~. Local application of low doses of NMDA into the striatum or nucleus accumhens has also proved ineffective in altering DA release

using this technique 7a9. One possible explanation of these divergent results is that the dialysis technique is based on a different sampling procedure (ex vivo determination of species by electrochemical detection) which works at a slower sampling rate than in vivo voltammetry. Consequently, the short-lived increase in DA release indicated in the present study might not be adequately reproduced by the dialysis technique. It is also possible that different routes of intracerebral administration of NMDA (e.g. via cannula or via dialysis probe) can influence these results. In this respect it should be noted that a certain time delay was always observed between the start of the i.c.v, infusion and the occurrence of the responses in nucleus accumbens. This may indicate diffusion of NMDA to a remote site but it may also indicate that NMDA-evoked an effect that propagated slowly through brain tissue. It is evident, however, that NMDA exerted a receptormediated influence on DA release in nucleus accumbens since the effect on Peak 2 was fully antagonized by infusion of the competitive NMDA receptor antagonist, CPP. Previous studies in anaesthetized rats have shown that the non-competitive NMDA receptor antagonist, dizocilpine, also is effective in this respect (unpublished observations). A phenomenon which should be considered as a possible explanation of these effects is spreading depression (SD). This phenomenon was first described by Leao 24 and is defined as a transient depression of neuronal activity that successively spreads to adjacent tissue s. Spreading velocity is slow (a few mm/min) and the event is accompanied by a large increase in extracellular K + concentration and a negative shift in field potential 5,2s. A concomitant influx of Ca 2+ ions has also been observed which indicates that SD triggers the release of neurotransmitters ts. In line with this assumption, a pronounced release of DA has been shown in the striatum during SD 27. Little is known about the cause of these abrupt changes in neuronal function. It should be noted, however, that excitatory amino acids are thought to be involved in the generation of SD and that NMDA has been shown to initiate SD in vitro z~. Several observations in the present study are consistent with the nature of SD. The time delay between the initiation of the NMDA infusion and the appearance of the responses in nucleus accumbens corresponds well to the spreading velocity of SD. The time-course of the changes in release of DA and the estimated peak increase in concentration of this species also corresponds well to what has previously been observed during SD 27. Furthermore, successive infusions of NMDA at short time intervals failed to evoke new

2(10 responses and continuous infusion of NMDA for a longer time period did not change the shape of these responses suggesting that they were not dependent on the dose infused within a broad dose interval. Previous studies using trans-striatal dialysis technique have shown that higher doses of NMDA locally infused into the rat striatum produce a pronounced release of DA in this area :. Co-infusion of atropine fully antagonized this effect suggesting that tile effect of NMDA on striatal DA release is mediated via a cholinergic neuron. Since the muscarinic receptor antagonist, atropine, was unable to block the effects of NMDA in the present study no evidence is provided for a similar mechanism mediating NMDA-evoked release of DA in nucleus accumbens. Neither co-infusion nor systemic administration (2 mg/kg, i.p., unpublished observations) of this drug had any effects on the ~MDA-evoked increases in peak heights recorded. The rapid increase and decay in DA release might hlternatively be explained by a dual action of NMDA on neuronal activity in nucleus accumbens. Studies by 12heramy and co-workers ~ indicate that DA release in the caudate nucleus is under presynaptic glutamatergic control. A biphasic effect was observed. Low doses of GLU locally applied in the caudate nucleus of anaesthetized cats stimulated DA release and high doses reduced the release, it was suggested that the increased release was caused by a direct stimulatory action of GLU on receptors located on dopaminergic terminals while the inhibitory effect was mediated, at least in part, by activation of less sensitive GLU receptors located on inhibitory GABA-ergic neurons. It is not clear from these studies whether the NMDA subtype of GLU receptors is involved in these processes. It is possible, however, that dopaminergic neurons in the nticleus accumbens are under a similar glutamatergic control. Initially, NMDA infusion caused an increase in DA release, possibly mediated by a direct stimulation of GLU receptors located on dopaminergic terminals, Continued infusion of NMDA into the intracerebroventricular space could then have caused an increasing activation of less sensitive GLU receptors located on inhibitory GABA-ergic neurons. This would explain the short latency reversal of the DA release. However, co-infusion of the GABA receptor antagonist, picrotoxin, failed to modify the responses to NMDA which makes this explanation less probable. The NMDA-evoked effect also included changes in Peak I indicating an increased release of AA. The functional significance of extracellular AA is largely unknown but this topic has received an increasing interest. Resent studies indicate that AA may influence central neuronal function and act as a neuromod-

ulator in the neostriatum 1. Changes in extracellular AA have also been suggested to signal the uptake of newly released GLU and hence serve as an index of GLU release 12. In line with this suggestion, extracellular AA show rapid changes in response both to local infusions of GLU into various brain regions ~6 and to behavioural activation 3. The AA recordings performed may consequently imply that NMDA evoked an increase in the release of GLU in nucleus accumbens. Indeed, NMDA has previously been shown to release GLU in the rat striatum in vivo 4(i. In this respect it is interesting to note that release of GLU has been suggested to be an important link in a chain of chemical events responsible for the initiation of SD 2,,37. Another possible explanation of the increase in extracellular AA concentration might be derived from the fact that this species, being an antioxidant, also seems to serve an important scavenger function in the brain 3~. DA is readily oxidized in two steps to its corresponding quinone which is a highly reactive molecule. The electrochemical oxidation of DA may result in a significant increase in the formation of DA-quinone and chemically related free radicals. This might in turn lead to increased scavenging activity reflected by the increase in AA aimed at protecting the brain against these potentially cytotoxic species ~4. Various behavioural effects have been reported to be evoked by intracerebral administration of NMDA ~,1~.21.2~.in the present study clear evidence of behavioural activation was observed. This was mainly reflected in an increased locomotor activity consisting of a well coordinated movement pattern occasionally interrupted by grooming, rearing or jumping behaviour. The initiation of the behavioural effects of NMDA coincided in time with the increase in release of DA recorded in the nucleus accumbens (to be published). This area receives dopaminergic neurons from the ventral tegmental area and these fibres have been associated with the regulation of motor function 2~,. Local application of DA in the nucleus accumbens causes locomotor stimulation and lesion of the mesolimbic DA system inhibits locomotor activity 2o.22,3.~.36. The increase in extracellular concentration of DA accompanying the locomotor stimulatory effect supports an involvement of DA in this effect. It should be emphasized, however, that there is evidence that locomotor activity can be stimulated via mechanisms independent of a stimulation of DA release in nucleus accumbells 2,t,,2,~,

In conclusion i.c,v, administration of a low dose of NMDA was found to evoke transient changes in the release of oxidizable species including catecholamines in nucleus accumbens, A possible explanation of these

201 effects is an initiation of SD. The rapid time-course of the effects, the pronounced release of species indicated and the low dose of NMDA involved call for careful consideration of these effects when interpreting the results of experiments where stimulation of glutamatergic mechanisms are involved. Acknowledgements. Mr. Stellan Ahl is gratefully ackaowledged for the construction of the electrode implantation assembly. This work was supported by grants from the Swedish Medical Research Council (4247 and 9075), R'~dman Ernst Coilianders Stiftelse, Gyllenstiernska Krapperups Stiftelse, Stiftelsen Lars Hiertas Minne, Wilhelm och Martina Lundgrens Vetenskapsfond, ~the Scandinavian Societ.y for Psychopharmacology, the Swedish Alcohol Monopoly Foundation for Alcohol Research and the Swedish Society of Medicine and Torsten och Ragnar S6derbergs Stiftelse.

REFERENCES ! Basse-Tomusk, A. and Rebec, G.V., Corticostriatal and thalamic

regulation of amphetamine-induced ascorbate release in the neostriatum, Pharmacol. Biochem. Behat'., 35 (I 990) 55-60. 2 Boldry, R.C. and Uretsky, N.J., The importance of dopaminergic neurotransmission in the hypermotility response produced by the administration of N-methyI-D-aspartic acid into the nucleus accumbens, Neuropharmacology, 27 (1988) 569-577, 3 Boutelle, M,G,, Svensson, L, and Fillenz, M,, Rapid changes in striatal ascorbate in response to tail-pinch monitored by constant potential voltammetry, Neuroscienee, 30 (I 989) I I - 17. 4 Buda, M., Gonon, F,, Cespuglio, R., Jouvet, M. and Pujol, J.F., in rive electrochemical detection of catechols in several dopaminergic brain r¢gions of anaesthetized rats, Eur. J. Pharmacol., 73 (1981) 61-68, 5 Burel, J., Bure,~ova, O. and Kirvanek, J., The Mechanism and Applications of Leao's Spreading Depression of Electroencephalographic Acti~'ity, Academic, New York, 1974. 6 Carlsson, M, and Carlsson, A,, The NMDA antagonist MK.801 causes marked locomotor stimulation in monoamine.depleted mice, J. Neural Transm., 75 (1989) 221-226. 7 Carter, C J,, L'Heureux, R. and Scatton, B., Differential control by N, methyl-i)-aspartate and kainate of striatal dopamin~ release in rive: a trans-striatal dialysis study, J. Neurochem., 51 (1988) 462-468, 8 Cheramy, A,, Rome, R., Oodeheu, G., Baruch, P. and Glowinski, J., in rive presynaptic control of dopamine release in the cat caudate nucleus - - II. FaciUtatory or inhibitory influence of t.-glutamate, Neuroscience, 19 (1986) 1081-1090. 9 Christie, MJ., Summers, RJ., Stephenson, J.A., Cook, C.J., Beart, P.M., Excitatory amino acid projections to the nucleus accumbens septi in the rat: a retrograde transport study utilizing D['~H] aspartate and ['~H]GABA, Neuroscience, 27. ~19~71 425-439. 10 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons, Acta Physiol. Scand., 62 (Suppl. 232) (1964) 1-55. II Engel, J.A,, Fahlke, C., Hulthe, P., H~rd, IF.., Johannessen, K., Snape, B. and Svensson, L., Biochemical and behavioral eviderxe for an interaction between ethanol and calcium channel antagonists, J. Neural Transm., 74 (1988) 181-193. 12 Fillenz, M., O'Neill, R.D. and Grunewald, R.A., Changes in extraceilular brain ascorbate concentration as an index of excitatory amino acid release, in M.H. Joseph, M. Fillenz, I.A. Macdonald and C.A. Marsden (Eds.), Monitoring Neurotransmitter Release During Behaviour, Ellis Horwood Health Science Series, Chichester, 1986, pp. 144-163. 13 Friedmann, M. and Gerhardt, G.A., In rive electrochemical studies of the effects of N-methyI-D-aspartate (NMDA) on dopamine nerve terminals in the neostriatum of the anesthetized rat, Prec. West. Pharmacol. Soc., 32 (19891 143-147. 14 Fornstedt, B. and Carlsson, A., Vitamin C deficiency facilitates

5-S-c}steinyl-dopamine formation in guinea pig striatum, J. Ne~,rachem,, 56 (1991) 407-414. 15 Hansen, A,J. and Zeuthen, T., Extracellular ion concentrations during spreading depression and ishemia in the rat brain cortex, Acta Physiol. Scand., 113 (1981)437-445. 16 Ghasemzadeh, B., Cammack, J. and Adams, R.N., Dynamic changes in extracellular fluid ascorbic acid monitored by in vivo electrochemistry, Brain Res., 547 (19911 162-166. 17 Gonon, F.G., Fombariet, C.M., Buda, M.J. and Pujol, F.J., Electrochemical treatment of pyrolytic carbon fiber electrodes, Anal, Chem., 53 (1981) 1386-1389. 18 Grace, A.A., Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: a hypothesis for the etiology of schizophrenia, Neuroscience, 41 (19911 1-24. 19 Imperato, A., HonorS, T. and Jensen, L.H., Dopamine release in the nucleus caudatus and in the nucleus accumbens is under glutamatergic control through non-NMDA receptors: a study in freely-moving rats, Brain Res., 530 (19901 223-228. 20 Jackson, D.M., And~n, N.E. and Dahlstr6m, A., A functional effect of dopamine in the nucleus accumbens and in some other dopamine-rich parts of the rat brain, Psychopharmacoiogia, 45 (19751 139-149. 21 Kabuto, H., Yokoi, l., Mizukawa, K. and Mori, A., Effects of an N-methyI-D-aspartate receptor agonist and its antagonist CPP on the levels of dopamine and serotonin metabolites in rat striatum collected in vivo by using a brain dialysis technique, Neurochem. Res., 14 (19891 1075-1080. 22 Koob, G.F., Stinus, L. and Le Meal, M., Hyperactivity and hypoactivity produced by lesions to the mesolimbic dopamine system, Behat,. Brain Res., 3 (19811 341-351. 23 Lauritzen, M., Rice, M.E., Okada, Y. and Nicholson, C., Quisqualate, kainate and NMDA can initiate spreading depression in the turtle cerebellum, Brain Res., 475 (19881 317-327. 24 Leao, A.A.P., Spreading depression of activity in the cerebral cortex, J. Neurophysiol,, 7 (19441 359-390. 25 Mogenson, G.J., Jones, D.L. and Yim, C.Y., From motivation to action: functional interface between the limbic system and the motor system, Prog. Neurobiol,, 14 (1980) 69-77. 26 Mogenson, G.J., Yang, C.R. and Yim, C.Y., Influence of dopamine on limbic inputs to the nucleus accumbens, in P.W. Kalivas and C.B, Nemeroff (Eds.), The Mesocorticol#nbic Dopamine System, VoL 537, Ann. N.Y. Acad. Sci., New York, 1988, pp. 86-100. 27 Moghaddam, B., Schenk, J.O., Stewart, W.B. and Hansen, A.J., Temporal relationship between neurotransmitter release and ion flux during spreading depression and anoxia, Can. J. Physiol. Pharmacol,, 65 (1987) ! 105- I 110. 28 Nieholson, C. and Kraig, R.P., The behavior of extracellular ions during spreading depression, in T. Zeuthen (Ed.), The Applica. tion of Ion.Selective Microelectrodes, Elsevier, Amsterdam, 1981, pp. 217-238. 29 O'Neill, K.A., Carelli, R.M., Jarvis, M.F. and Liebman, J.M., Hyperactivity induced by N.methyI.D.aspartate injections into nucleus accumbens: lack of evidence for mediation by dopaminergic neurons, Pharmacol. Biochem. Behav., 34 (1989) 739-745. 30 Osteryoung, J.G. and Osteryoung, R'.A., Square wave voltammetry, Anal Chem., 57 (1985) 101-110. 31 Padh, H., Cellular functions of ascorbic acid, Biochem. Cell. Blot,, 68 (19901 i 166-1173. 32 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, Sydney, 1986. 33 Pijenburg, A.J.J. and van Rossum, J.M., Stimulation of locomotor activity following injection of dopamine into the nucleus accumbens, J. Pharm. Pharmacol., 25 (1973) 1003-1005. 34 Roberts, D.C.S. and Zito, K.A., Interpretation of lesion effects on stimulant self.administration, in M.A. Bozarth (Ed.), Methods of Assessing the Reinforcing Properties of Abused Drugs, Springer, New York, 1987 pp. 87-104. 35 Stellar, J.R., Kelley, A.E. and Corbett, D., Effects of peripheral and central dopamine blockade on lateral hypothalamic selfstimulation: evidence for both reward and motor deficits, Pharmacol. Biochem. Behav., 18 (1983) 433-442.

202 36 Svensson, L. and Ahlenius, S., Suppression of exploratory locomotor activity by the local application of dopamine or ~noradrenaline to the nucleus accumbens of the rat, Pharmacoi. Biochem. Behar.o 19 (1983) 693-699. 37 Van Harreveld, A., Two mechanisms for spreading depression in the chicken retina, J. Neurobiol., 9 (1978) 419-431. 38 Walaas, !. and Fonnum, F., Biochemical evidence for glutamate as a transmitter in hippocampal efferents to the basal forebrain

and hypothalamus in the rat brain, Neuroscience, 5 (1980) 16911698. 39 Wise, R.A. and Rompre, P.-P., Brain dopamine and reward, Ann. Rer. Psychol., 40 (1989) 191-225. 40 Young, A.M.J. and Bradford H.F., N-Methyl-D-aspartate releases excitatory amino acids in rat corpus striatum in rive, .i. Neurochem., 56 (1991) 1677-1683.