Brabz Research, 597 (1992) 138-143 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
138
BRES 18299
Unilateral neostriatal kainate, but not 6-OHDA, lesions block dopamine agonist-induced ascorbate release in the neostriatum of freely moving rats R.C. Pierce, D.W. Miller, D.B. Reising a n d G . V . R e b e c Program in Neural Science, Department of Psychology, Indiana Unirersity, Bloomington, IN 47405 (USA) (Accepted 30 June 1992)
Key words: Amphetamine; Ascorbate; GBR-12909; 6-Hydroxydopamine; Kainate, Neostriatum; Quinpirole; SKF-38393; Voltammetry
Unilateral kainate lesions of the neostriatum and 6-hydroxydopamine (6-OHDA) lesions of the medial forebrain bundle were used to assess the role of neostriatal and ascending dopaminergic neurons, respectively, on dopamine-agonist induced release of neostriatal ascorbate as measured voltammetrically in freely moving rats. Electrochemically modified, carbon-fiber electrodes recorded the effects of direct (a combination of 10 mg/kg SKF-38393 and 1.0 mg/kg quinpirole) as well as indirect (2.5 mg/kg D-amphetamine or 20.0 mg/kg GBR-12909) dopamine agonists. Relative to controls, kainate, but not 6-OHDA, lesions abolished the ability of both direct and indirect dopamine agonists to induce neostriatal ascorbate release. These results suggest that unlike dopaminergic afferents, neostriatal output pathways play a critical role in the modulation of neostriatai ascorbate levels.
INTRODUCTION Steadily accumulating evidence suggests that the water-soluble vitamin, ascorbate, acts in the neostriat-m to modulate dopamine-mediated behavior. Ascorbate infusions directly into this structure, for example, attenuate the behavioral effects of amphetamine, an indirect dopamine agonist 3~, while systemic administration of ascorbate potentiates the behavioral effects of the dopamine antagonist, haloperido124. Furthermore, repeated coadministrations of ascorbate and haloperidol potentiate the supersensitive behavioral response to the direct dopamine agonist, apomorphine, typically observed following chronic treatment with haloperidol alone e~. Taken together, these results suggest a neuroleptic-like action of neostriatal ascorbate. Despite an apparent interaction between ascorbate and dopamine, the neuronal systems underlying ascorbate release in the neostriatum remain unclear. Although amphetamine and other dopamine agonists increase the extracellular level of neostriatal ascorbate and this effect is reversed by dopamine
antagonists ~6'2°'35, direct infusions of amphetamine into the neostriatum fail to elevate ascorbate levels in this brain region and may, in fact, cause a slight reduction 34. In contrast, amphetamine infusions into the substantia nigra significantly increase neostriatal ascorbate release, suggesting that amphetamine and perhaps other dopamine agonists alter neostriatal ascorbate release by acting outside the neostriatum 33,34. This view is consistent with evidence that destruction of dopaminergic terminals in the neostriatum fails to block amphetamine-induced ascorbate release 6.Hu2. The corticostriatal system has received the most attention as the source of extraceilular ascorbate in the neostriatum largely because lesions of this pathway significantly attenuate the basal level of this vitamin 2'~7. It is important to note, however, that relative to basal levels, amphetamine-induced ascorbate release is actually enhanced following such lesions, suggesting that other mechanisms also participate in this process 2. One likely participant is the neostriatal projection to the substantia nigra. When this system is damaged by electrolytic lesions of the crus cerebri, amphetamine-
Correspondence: G.V. Rebec, Program in Neural Science, Department of Psychology, Indiana University, Bloomington, IN 47405, USA. Fax: (1)
(812) 855-4520.
139 induced ascorbate release in the neostriatum is completely abolished 33. The crus cerebri, however, contain descending fibers from several forebrain structures, including cerebral cortex, making it difficult to specify the exact source of ascorbate release. In the present report, we combined voltammetric techniques with neurotoxic lesions of selective neuronal pathways to shed further light on the systems underlying the release of neostriatal ascorbate produced by amphetamine and other dopamine agonists. One group of animals received unilateral injections of 6-hydroxydopamine (6-OHDA) into the medial forebrain bundle (MFB) to destroy all ascending dopaminergic systems, including those projecting to the neostriatum. Separate animals sustained large-scale destruction of the neostriatum via unilateral infusions of kainate. Our results indicate that mechanisms intrinsic to the neostriatum, rather than brainstem dopaminergic projections, regulate the increase in neostriatal ascorbate produced by both direct and indirect dopamine agonists. MATERIALS AND METHODS
Animals Adult, male, Sprague-Dawley rats (350-450 g), bred in our animal colony, were used as subjects. They were housed individually under standard laboratory conditions with food and water available continuously.
Kainate and 6.0HDA lesions Prior to surgery, all animals were anesthetized with Equithesin (0.33 ml/kg, i.p.); diazepam (10 mg/kg; i.p.) was administered to kainate animals to minimize seizures while 6-OHDA animals were pretreated with the monoamine oxidase inhibitor, pargyline (40 mg/kgl i.p.), and the norepinephrine re-uptake blocker, desipramine (25 mg/kg; i.p.). The animals then were placed in a stereotaxic instrument and a hole was drilled unilaterally through the skull over the neostriatum (1.0 mm anterior and 2.5 mm lateral to bregmat~). For kainate lesions, a cannula was lowered into the neostriatum (5.0 mm ventral to the dura) and 1.0/zl (2.0 g g / t t l ) kainate was infused over a 5-rain period. For 6-OHDA lesions, a second hole was drilled over the MFB (4.0 mm posterior and 2.0 mm lateral to bregma), a cannula was lowered through this hole into the MFB (8 mm ventral to the dura) and 2.0/zl (4.0 tt~;//zl) 6-OIIDA was infused over a 5-rain period. After the infusion (kainate or 6-OHDA), the cannula was left in place for 5 rain. The cannula then was removed and a plastic cylindrical hub, designed to mate with an electrode holder for voltammetric recordings, was centered over the hole located above the neostriatum and cemented onto the skull. Prior to cementing, the hole over the MFB was filled with Gelfoam, an absorbable gelatin sponge. Control animals received the same treatment as lesion animals, including lowering the cannula into the brain, but no neurotoxin was administered. Following surgery, all animals were treated with 0.20 ml (25 rag) Rocephin (Roche) to minimize infection. Voltammetric recordings were performed after a recovery period of approximately 7 days.
Voltammetry Microvoltammetric electrodes were prepared from Thorneli P-55 carbon fibers (Union Carbide) as described previously7'12. The active
electrode surface is 200 ~ m in length with a 5-/~m radius. Each electrode was treated in a citrate-phosphate buffer solution by application of a triangular wave potential (70 Hz, 0.09-3.0 V vs. a saturated calomel reference electrode, SCE) for 20 s followed by 10 s of 0.5 V cathodal current, 3 s of 0.9 V anodal current, and 8 s of 1.5 V cathodal current. This treatment modifies the active surface of the electrode to allow discrimination of ascorbate from catechols and other easily oxidizable compounds 1°. Each electrode then was tested with staircase voltammetry in 100/zM ascorbate and 20/zM 3,4-dihydroxyphenylacetic acid (DOPAC) until the response stabilized. For in vivo voltammetry, the electrode was placed into a holder consisting of a stainless steel, luer-lock fitting that mates with the hub placed on the animal's skull during surgery. The holder is equipped with a threaded assembly that allows the electrode to be raised and lowered through the brain. For the experiments in this report, the electrode was lowered 5 mm ventral to the brain surface into the neostriatum. Generation of waveforms for staircase voltammetry and storage of sampled current was performed by a computer (IBM-XT) interfaced to a locally constructed potentiostat of conventional 3-electrode design6. The carbon-fiber electrode served as the working electrode, while the stainless steel electrode holder, which extends 2.5 mm ventral to the brain surface, served as both the reference and auxiliary electrodes. A potential was applied to the working electrode in 2.44-mV steps from 0 to 500 mV vs reference to insure the oxidation of ascorbate and DOPAC. Short pulses (98 ms) were used in order to restrict sampling to the solution pool near the electrode surface. The scan rate was set at 10 mV/s; scans were obtained at 2-min intervals. Voitammograms displayed peaks representing ascorbate (approximately 150 mV vs. reference) and DOPAC (approximately 300 mV vs. reference), as well as a third more positive peak thought to represent 5-hydroxyindolacetic acid and uric acid 4.5. Drug-induced changes in extracellular ascorbate were tested after a stable baseline was established and all animals were resting quietly. Separate groups of both lesion and control animals received either 2.5 mg/kg D-amphetamine (s.c.), 20.0 mg/kg GBR-12909 (s.c.) or the combination of 10.0 mg/kg SKF-38393 (i.p.) and 1.0 mg/kg quinpirole (s.c.). These doses of dopamine agonists are known to produce reliable increases in neostriatal ascorbate as well as characteristic patterns of locomotion and stereotyped behavior (see ref. 19). In all cases, voltammetric signals were recorded for another 50 min. When voltammetric recording was completed, all animals were anesthetized with sodium pentobarbital. Some of the working electrodes used for each drug group were removed for postcalibration in 100 /zM ascorbate and 20 ~M DOPAC. In order to verify the recording site, current was passed through the electrodes that remained in the brain in order to produce a small lesion. Following transcardial perfusion with normal saline and 10% formalin, brains were removed from the lesioned animals. Brain tissue was subsequently frozen, sectioned, and stained with Cresyl violet. For histological analysis, sections were observed microscopically (Carl Zeiss microscope; 16× and 40x objectives) for electrode-placement lesions and, when appropriate, extent of damage caused by kainate infusions. The extent of kainate lesions was determined by extent of gliosis and cell death in the neostriatum and surrounding structures. An acceptable kainate lesion was operationally defined as gliosis and cell death in at least 75% of the neostriatum with little or no damage outside of this structure (with the exception of cortical damage dorsal to the neostriatum, see Results).
Drugs D-Amphetamine sulfate (Sigma), quinpiroie (RBI), pargyline (Sigma), and kainate (Sigma) were mixed in 0.9% NaCI. GBR-12909 (RBI), SKF-38393 (RBI) and desipramine (Sigma) were mixed in 5% tartaric acid. 6-OHDA (Sigma) was mixed in a 0.9% NaCI-0.2% ascorbate solution. Diazepam (Sigma) was mixed in a vehicle containing 50% saline, 40% propylene glycol and 10% ethanol. Amphetamine and 6-OHDA were mixed as the free base, all other compounds were expressed as the salt.
140 Statistics
TABLE I
Voltammograms were expressed as percent baseline following drug treatment. Between-subjects analyses of variance were performed on all our data. Further pairwise comparisons were performed with Tukey's HSD.
Vohammetrically recorded extraceihdar levels (i~M +_S.E.M) of basal ascorbate and catechols in the neostriatum of control, kainate- and 6-OHDA-lesioned rats
The asterisk represents a significant difference from the control and kainate groups (P < 0.05, Tukey's HSD).
RESULTS All identifying lesions were located in anterior neostriatum: 4.0 to 5.0 mm ventral to the brain surface and approximately 2.5 mm lateral to the midline. As shown in Table I, based on postcalibration values, baseline extracellular neostriatal ascorbate levels were similar across treatment groups. These values are comparable with neostriatal ascorbate levels reported in previous studies3.10,27,32. Kainate a n d 6 - O H D A lesion verification
The effectiveness of kainate and 6 - O H D A lesions were verified histologically and biochemically, respectively. Fig. 1 depicts the typical extent of gliosis and cell death following a unilateral kainate lesion. In all cases, approximately 75% of the neostriatum was damaged. Table I includes voltammetric recordings not only of the ascorbate peak but also the catechol peak, which represents both dopamine and its metabolite, DOPAC. Animals in the control and kainate lesioned
Treatment
n
Ascorbate
Catechols
Control Kainate 6-OHDA
10 6 6
314 (24) 238 (49) 306 (38)
24.1 (2.2) 22.5 (3.0) 2.3 (1.5) *
groups had similar catechol oxidation peaks; these values are comparable with those reported in previous studies g'27. As expected, basal catechol levels decreased dramatically following destruction of neostriatal dopaminergic afferents via 6-OHDA. Indeed, the catechol values in the 6 - O H D A group represent a reduction of more than 95% relative to controls. Dopamine agonist-induced behaviors in kainate- and 6-OHDA- as well as sham-lesioned animals also were noted. The sham-lesioned animals displayed behaviors typically observed following administration of amphetamine and GBR-12909 (locomotion, sniffing, rearing and head bobbing) as well as the combined administration of SKF-38393 and quinpirole (sniffing, head
Fig. I. Photomicrograph of a coronal section (approximately 1.0 mm anterior to bregma) through the neostriatum of a kainate-lesioned rat brain showing typical extent of damage. The darkly stained areas (left side of the section) represent the cell death and gliosis resulting from the lesion. Note damage to approximately 75% of the neostriatum. Although some collateral damage occurred in the cortex dor-sal to the neostriatum, cortical damage alone had no effect on ascorbate release (see Results).
141 bobbing, biting and licking). Similar to previous resuits 2°, the onset of these behaviors was within 10 rain of drug administration and persisted for the entire observation period (50 min). The lesioned animals displayed behaviors similar to the sham-lesioned animals with the addition of circling, which was particularly intense in the 5-OHDA groups. This is not surprising since dopamine agonist-induced circling ipsilateral to a 6-OHDA-lesion is associated with unilateral striatal dopamine depletions in excess of 950~ 14'37. Similarly, unilateral striatal kainate lesions are known t6 produce circling contralateral to the lesioned side a°. While sham-lesioned SKF-38393-quinpirole animals displayed little locomotion, 6-OHDA- and kainate-lesioned animals circled to a similar extent regardless of the dopamine agonist administered (data not shown).
Effects of kainate and 6-OHDA lesions on dopamine agonist-induced ascorbate release Statistical analyses revealed no significant difference in extracellular ascorbate levels between both sham-lesioned groups 50 min after administration of either a m p h e t a m i n e ( t 2 = 0.83, n.s.) or vehicle (t 2 = 0.6, n.s.). In both cases, amphetamine caused at least a 50% increase in ascorbate, whereas ascorbate declined slightly following vehicle treatment. Thus, the respective amphetamine and vehicle responses of each shamlesioned group was combined for comparison with both kainate- and 6-OHDA-lesioned animals. The ControlAmph group in Fig. 2 shows the time course of the neostriatal ascorbate response following administration of 2.5 mg/kg D-amphetamine. This gradual increase in
0~0 &mA
ControI-Amph 60HDA-Amph
C l u E ] Koinote-Amph O m O Control-Vehicle
80. bJ u3 +1 q~ ,JC
40.
"~
20.
rn
O.
~
-20.
fj,,}
j6
60.
13- -40
I
I
I
I
I
10
20
30
40
50
Minutes P o s t - I n j e c t i o n
Fig. 2. Effects of 2.5 mg/kg D-amphetamine on ascorbate release in the neostriatum of kainate- (Kainate-Amph), 6-OHDA- (6-OHDAAmph) and sham-lesioned (ControI-Amph) animals as well as the effects of vehicle treatment in sham-lesioned (Control-Vehicle) animals. All responses are presented as the mean (+_S.E.) percent predrug baseline for each 10-min period following amphetamine or vehicle administration. Note the steady increase in extracellular ascorbate in both the ControI-Amph and 6-OHDA-Amph groups. Conversely, ascorbate in the Kainate-Amph group did not differ from the Control-Vehicle group.
Koinote .-.
80
~
40
~
20
"g
o
~
GBR
60HDA
~
Control
SKF-Quin
Fig. 3. Effects of direct (a combination of 10.0 mg/kg SKF-38393 and 1.0 mg/kg quinpirole) and indirect (20.0 mg/kg GBR-12909) dopamine agonists on neostriatai ascorbate in kainate, 6-OHDA, and control (sham-lesioned) animals. All responses are displayed as the mean (_+ S.E.) percent predrug baseline for the 6-min period 44-59 min following drug administration. Note the similarities between the responses of these agonists to that of amphetamine (see Fig. 2). Thus, whereas both direct and indirect dopamine agonists induce neostriatal ascorbate release in control and 6-OHDA animals, the kainate group shows a decrease from baseline. The * represents a significant difference from the kainate group.
ascorbate over the first 50 min of the drug response is typically observed following the administration of both indirect (amphetamine and GBR-12909) and direct (SKF-38393-quinpirole combination) dopamine agonists (see ref. 19). Also shown in Fig. 2 are the time courses for the effects of amphetamine in kainate- and 6-OHDA-lesioned animals, rhe overall ANOVA and subsequent pairwise comparisons showed no significant difference between kainate-lesioned and ControlVehicle animals. By contrast, there was a significant increase in ascorbate in the 6-OHDA-lesioned animals following amphetamine. Indeed, pairwise comparisons revealed that the ascorbate response of the 6-OHDA group was not significantly different from the ControlAmph group at any of the time points examined. Similar to the control groups outlined above, there were no significant differences between the sham kainate- and sham 6-OHDA-iesioned groups when administered either GBR-12909 ( t 2 = 2.26, n.s.) or the SKF.38393-quinpirole combination (t: = 0.68, n.s.) during voltammetric recordings. Therefore, these groups were combined into two control groups, respectively, for comparison to data obtained from lesioned animals. Results similar to those seen with amphetamine were observed following the administration of GBR-12909 and the SKF-38393-quinpirole combination, as shown in Fig. 3. Thus, while these agonists clearly increased neostriatal ascorbate 50 min after their administration in the Control and 6-OHDA groups, they had no effect on neostriatal ascorbate in kainate-lesioned animals. In addition to extensive neostriatal damage in the kainate group, there also was some damage to the
142 cortex dorsal to the neostriatum (see Fig. 1). To eliminate cortical damage as a factor underlying the inability of amphetamine to increase neostriatal ascorbate levels in this group, we inflicted discrete cortical lesions similar to those observed following neostriatal lesions in a separate group of animals (0.5 /.tg//.tl kainate unilaterally over 5 min; 1.0 mm anterior and 2.5 mm lateral to bregma, 2.0 mm ventral to dura). Voltammetric recordings 7 days following surgery revealed an increase (n = 2) in ascorbate similar to that observed in control animals (40% mean increase) 50 min after administration of 2.5 mg/kg o-amphetamine.
DISCUSSION Consistent with previous results, our data indicate that amphetamine ~'~L23 as well as other dopamine agonists ~9 increase behavioral activation as well as the level of ascorbate in the extracellular fluid of the neostriatum. Although the functional significance of this increase is not well understood, dopaminergic neurons appear to play little or no role in it. Thus, as reported for rats with 6-OHDA lesions of the neostriatum 6'~2, we found that the amphetamine-induced increase in neostriatal ascorbate release is not significantly attenuated by large-scale depletion of ncostriatal dopamine. Interestingly, however, we found, in agreement with previous evidence ~°, that dopamine depletions delayed somewhat the increase in ascorbate induced by amphetamine (see Fig. 2). Our results with 6-OHDA lesions also applied to OBR-12909 and to the SKF-38393-quinpirole combination, indicating that neither indirect nor direct dopamine agonists require an intact dopaminergic projection to the neostriatum in order to induce ascorbate release. Our results also confirm that ascorbate is not released from neostriatai dopaminergic terminals 6.~°a2. In fact, electrical stimulation of the MFB, which activates dopaminergic neurons ~a, does not induce neostriatal ascorbate release ~3'2s. Furthermore, neostriatal areas of high dopaminergic activity are inversely related to areas of high extracellular ascorbate 3. We used 6-OHDA lesions of the MFB, which are known to cause dramatic decreases in dopamine levels throughout the entire forebrain, not just the neostriaturn 3~''37. Thus, it is difficult to argue that residual dopamine in nucleus accumbens or other dopaminerich forebrain regions contributes to our results with either amphetamine or GBR-12909. Unless these drugs exert a postsynaptic dopamine-mimetic effect like SKF-38393 and quinpirole, which seems unlikely, amphetamine and GBR-12909 must act on non-dopamin-
ergic systems to promote ascorbate release in the neostriatum. Ample evidence appears to link ascorbate release with glutamate-containing neurons. Brain regions with high concentrations of glutamate, such as neostriatum, hippocampus, and cortex, also contain high levels of ascorbate s'29. Furthermore, glutamate infusions into each of these areas promotes ascorbate release saT. In addition, cortical lesions, which remove a major source of glutaminergic input to the neostriatum, significantly lower basal levels of ascorbate 2'!7. Collectively, these results indicate that ascorbate is released from glutaminergic axon terminals, but the mechanism by which both direct and indirect dopamine agonists influence this process remains to be established. Several lines of evidence suggest that neostriatal ascorbate release from corticostriatal terminals may be regulated by the substantia nigra, which influences cortical activity via relays in the thalamus. It is known, for example, that unilateral infusions of amphetamine into the substantia nigra cause a bilateral increase in neostriatal ascorbate release 33'34. Furthermore, similar to cortical lesions, bilateral lesions of the ventromedial thalamus, a main target of outputs from substantia nigra pars reticulata (SNR), lower both basal levels of ascorbate and the amount of ascorbate released by amphetamine in the neostriatum 2. Because amphetamine increases the firing rate of SNR neurons 25'26, this drug may alter thalamic and eventually cortical activity and in that way influences neostriatal ascorbate release. Activity in the SNR is strongly influenced by feedback neurons originating in the neostriatum 22. This pathway is largely destroyed by neostriatal kainate infusions, which also prevent ascorbate release by either indirect or direct dopamine agonists. These results are consistent with previous evidence that electrolytic lesions of the crus cerebri, which also destroy the neostriatonigral pathway, abolish the ability of amphetamine infused directly into the SNR to increasc neostriatal ascorbate release 33. Kainate lesions of the neostriatum, however, also destroy cholinergic interneurons, which may play a role in ascorbate release. For example, the cholinergic agonist, pilocarpinc, produces a dose-dependent increase in neostriatal ascorbate release, and this effect is reversed by scopolamine 15. In summary, our results support a growing body of evidence that ascorbate is not released from dopaminergic axon terminals nor are such neurons critical for the release of neostriatal ascorbate induced by either indirect or direct dopamine agonists. Rather, this process appears to be controlled by the neostriatum itself,
143 either via cholinergic interneurons or via~ projection neurons that modulate a circuit involving nigral, thalamic, and cortical neurons. Acknowledgements. This research was supported by NSF Grant BNS 91-12055. We also acknowledge the technical assistance of Paul Langley and the secretarial work of Faye Caylor. We also thank Dr. Dale Sengelaub for assistance with histological analysis.
REFERENCES 1 Adams, R.N., In t'it,o electrochemical measurements in the CNS, Prog. Neurobiol., 35 (1990) 297-311. 2 Basse-Tomusk, A. and Rebec, G.V., Corticostriatal and thalamic regulation of amphetamine-induced ascorbate release in the neostriatum, Pharmacol. Biochem. Behav., 35 (1990)55-60. 3 Basse-Tomusk, A. and Rebec, G.V., Regional distribution of ascorbate and 3,4-dihydroxyphenylacetic acid (DOPAC) in rat neostriatum, Brain Res., 538 (1991) 29-35. 4 Cespuglio, R., Faradji, H., Ponchon, J.-L., Riou, F., Buda, M., Gonon, F., Pujol, J.-F. and Jouvet, M., In vivo measurement by differential pulse voltammetry of extracellular 5-hydroxTindoleacetic acid in the rat brain, J. Physiol. 77 (1981) 327-332. 5 Crespi, F., Sharp, T., Maidement, N.T. and Marsden, C,A., Differential pulse voltammetry: simultaneous in vivo measurement of ascorbic acid, catechols and 5-hydroxyindoles in the rat striatum, Brain Res., 332 (1984) 135-138. 6 Ewing, A.G., Wightman, R.M. and Dayton, M.A., In vivo voltammetry with electrodes that discriminate between dopamine and ascorbate, Brain Res., 249 (1982) 361-370. 7 Ewing, A.G., Dayton, M.A. and Wightman, R.M., Pulse voltammetry with microvoltammetric electrodes, Anal. Chem., 53 (1981) 1842-1847. 8 Ghasemzedah, B., Cammack, J. and Adams, R.N., Dynamic changes in extracellular fluid ascorbic acid monitored by in vivo electrochemistry, Brain Res., 547 (1991) 162-166. 9 Gonon, F., Buda, M., Cespuglio, R., Jouvet, M. and Pujol, J.F., In vivo electrochemical detection of catechols in the neostriatum of anaesthetized rats: dopamine or DOPAC?, Nature, 286 (1980) 902-904. 10 Gonon, F., Buda, M., Cespuglio, R., Jouvet, M. and Pujol, J.F., Voltammetry in the striatum of chronic freely moving rats: detection of catechols and ascorbic acid, Brain Res., 223 (1981) 69-80. 11 Justice Jr., J.B., Introduction to in vivo voltammetry, in J.B. Justice Jr. (Ed.), Voltammetry in the Neurosciences, Principles, Methods, and Applications, Humana Press, Clifton, New Jersey, 1987, pp. 3-101. 12 Kamata, K., Wilson, R.L., AIIoway, K.D. and Rebec, G.V., Multiple amphetamine injections reduce the release of ascorbic acid in the neostriatum of the rat, Brain Res., 362 (1986) 331-338. 13 Kuhr, W., Wightman, R.M. and Rebec, G.V., Dopaminergic neurons: simultaneous measurements of dopamine release and single-unit activity during stimulation of the medial forebrain bundle, Brain Res., 418 (1987) 122-128. 14 Miller, R. and Beninger, R.J., On the interpretation of asymmetries of posture and locomotion produced with dopamine agonists in animals with unilateral depletions of striatal dopamine, Prog. Neurobiol., 36 (1991) 229-256. 15 Mueller, K., Voltammetric evidence in vivo of cholinergic modulation of extracellular ascorbic acid and uric acid in rat striatum, Brain Res., 408 (1987) 313-316. 16 Oh, C., Gardiner, T.W. and Rebec, G.V., Blockade of both DIand D2-dopamine receptors inhibits amphetamine-induced ascotbate release in neostriatum, Brain Res., 480 (1989) 184-189. 17 O'Neill, R.D., Fillenz, M., Sundstrom, L. and Rawlins, J.N.P., Voltammetrically monitored brain ascorbate as an index of excitatory amino acid release in the unrestrained rat, Neurosci. Lett., 52 (1984) 227-233.
18 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982. 19 Pierce, R.C. and Rebec, G.V., Stimulation of both DI and D2 receptors increases behavioral activation and ascorbate release in the neostriatum of freely moving rats, Eur. J. Pharmacol., 191 (1990) 295-302. 20 Pierce, R.C. and Rebec, G.V., Dopamine-, NMDA-, and sigmareceptor antagonists exert differential effects on basal and amphetamine-induced changes in neostriatai ascorbate and DOPAC in awake, behaving rats, Brain Res., in press. 21 Pierce, R.C., Rowlett, J.K., Bardo, M.T. and Rebec, G.V., Chronic ascorbate potentiates the effects of chronic haloperidol on behavioral supersensitivity but not D 2 dopamine receptor binding, Neuroscience, 45 (1991) 373-378. 22 Rebec, G.V., Electrophysiological pharmacology of amphetamine. In J. Marwah, (Ed.), Neurobiology of Drug Abuse, Monographs in Neural Sciences, Vol. 13, S. Karger, Basel, 1987, pp. 1-33. 23 Rebec, G.V., Changes in brain and behavior produced by amphetamine: a perspective based on microdialysis, voltammetry, and single-unit electrophysiology in freely moving animals. In R.R. Watson (Ed.), Biochemistry and Physiology of Substance Abuse, Vol. III, CRC Press, Boca Raton, FL, 1991, pp. 93-115. 24 Rebec, G.V., Centore, J.M., White, L.K. and Alloway, K.D., Ascorbic acid and the behavioral response to haloperidol: implications for the action of antipsychotic drugs, Science, 227 (1985) 438-440. 25 Rebec, G.V. and Groves, P.M., Apparent feedback from the caudate nucleus to the substantia nigra following amphetamine administration, Neuropharmacology, 14 (1975) 275-282. 26 Ruffieux, A. and Schultz, W., Dopaminergic activation of reticulata neurones in the substantia nigra, Nature, 285 (1980) 240-241. 27 Schenk, J.O., Miller, E., Gaddis, R. and Adams, R.N., Homeostatic control of ascorbate concentration in CNS extracellular fluid, Brain Res., 253 (1982) 353-356. 28 Stamford, J.A., Kruk, Z.L. and Millar, J., Ascorbic acid does not modulate stimulated dopamine release: in vivo voltammetric data in the rat, Neurosci. Lett., 60 (1985) 357-362. 29 Stamford, J.A., Kruk, Z.L. and Millar, J., Regional differences in extraceiiular ascorbic acid in the rat brain determined by high speed cyclic voltammetry, Brain Res., 299 (1984) 289-295. 30 Vecsei, L. and Beal, M.F., Comparative behavioral and neurochemical studies with striatal kainic acid- or quinolinic acid-lesioned rats, Pharmacol. Biochem. Behav., .~9 (1991)473-478. 31 White, L.K., Maurer, M., Kraft, M.E., Oh, C. and Rebec, G.V., lntrastriatal infusions of ascorbate antagonize the behavioral response to amphetamine, Pharmacol. Biochem. Behac., 36 (1990) 485-489. 32 Wightman, R.M., Brown, D.S., Kuhr, W.G. and Wilson, R.L., Molecular specificity of in vivo electrochemical measurements. In J. Justice (Ed.), Voltammetry in the Neurosciences, Humana Press, Clifton, NJ, 1987, pp. 103-138. 33 Wilson, R.L., Kamata, K., Bigelow, J.C., Rebec, G.V. and Wightman, R.M., Crus cerebri lesions abolish amphetamine-induced ascorbate release in the rat neostriatum, Brain Res., 370 (1986) 393-396. 34 Wilson, R.L and Wightman, R.M., Systemic and nigral application of amphetamine both cause an increase in extracellular concentration of ascorbate in the caudate nucleus of the rat, Brain Res., 339 (1985) 219-226. 35 Yount, S.E., Kraft, M.E., Pierce, R.C., Langley, P.E. and Rebec, G.V., Acute and long-term treatments alter extracellular ascorbate in neostriatum but not nucleus accumbens of freely moving rats, Life Sci., 49 (1991) 1237-1244. 36 Zahm, D.S., Compartments in rat dorsal and ventral striatum revealed following injection of 6-hydroxydopamine into the ventral mesencephalon, Brain Res., 552 (1991) 164-169. 37 Zigmond, M.J., Abercrombie, E.D., Berger, T.W., Grace, A.A. and Stricker, E.M., Compensation after lesions of central dopaminergic neurons: some clinical and basic implications, Trends Neurosci., 13 (1990) 290-295.
138
BRES 18299
Unilateral neostriatal kainate, but not 6-OHDA, lesions block dopamine agonist-induced ascorbate release in the neostriatum of freely moving rats R.C. Pierce, D.W. Miller, D.B. Reising a n d G . V . R e b e c Program in Neural Science, Department of Psychology, Indiana Unirersity, Bloomington, IN 47405 (USA) (Accepted 30 June 1992)
Key words: Amphetamine; Ascorbate; GBR-12909; 6-Hydroxydopamine; Kainate, Neostriatum; Quinpirole; SKF-38393; Voltammetry
Unilateral kainate lesions of the neostriatum and 6-hydroxydopamine (6-OHDA) lesions of the medial forebrain bundle were used to assess the role of neostriatal and ascending dopaminergic neurons, respectively, on dopamine-agonist induced release of neostriatal ascorbate as measured voltammetrically in freely moving rats. Electrochemically modified, carbon-fiber electrodes recorded the effects of direct (a combination of 10 mg/kg SKF-38393 and 1.0 mg/kg quinpirole) as well as indirect (2.5 mg/kg D-amphetamine or 20.0 mg/kg GBR-12909) dopamine agonists. Relative to controls, kainate, but not 6-OHDA, lesions abolished the ability of both direct and indirect dopamine agonists to induce neostriatal ascorbate release. These results suggest that unlike dopaminergic afferents, neostriatal output pathways play a critical role in the modulation of neostriatai ascorbate levels.
INTRODUCTION Steadily accumulating evidence suggests that the water-soluble vitamin, ascorbate, acts in the neostriat-m to modulate dopamine-mediated behavior. Ascorbate infusions directly into this structure, for example, attenuate the behavioral effects of amphetamine, an indirect dopamine agonist 3~, while systemic administration of ascorbate potentiates the behavioral effects of the dopamine antagonist, haloperido124. Furthermore, repeated coadministrations of ascorbate and haloperidol potentiate the supersensitive behavioral response to the direct dopamine agonist, apomorphine, typically observed following chronic treatment with haloperidol alone e~. Taken together, these results suggest a neuroleptic-like action of neostriatal ascorbate. Despite an apparent interaction between ascorbate and dopamine, the neuronal systems underlying ascorbate release in the neostriatum remain unclear. Although amphetamine and other dopamine agonists increase the extracellular level of neostriatal ascorbate and this effect is reversed by dopamine
antagonists ~6'2°'35, direct infusions of amphetamine into the neostriatum fail to elevate ascorbate levels in this brain region and may, in fact, cause a slight reduction 34. In contrast, amphetamine infusions into the substantia nigra significantly increase neostriatal ascorbate release, suggesting that amphetamine and perhaps other dopamine agonists alter neostriatal ascorbate release by acting outside the neostriatum 33,34. This view is consistent with evidence that destruction of dopaminergic terminals in the neostriatum fails to block amphetamine-induced ascorbate release 6.Hu2. The corticostriatal system has received the most attention as the source of extraceilular ascorbate in the neostriatum largely because lesions of this pathway significantly attenuate the basal level of this vitamin 2'~7. It is important to note, however, that relative to basal levels, amphetamine-induced ascorbate release is actually enhanced following such lesions, suggesting that other mechanisms also participate in this process 2. One likely participant is the neostriatal projection to the substantia nigra. When this system is damaged by electrolytic lesions of the crus cerebri, amphetamine-
Correspondence: G.V. Rebec, Program in Neural Science, Department of Psychology, Indiana University, Bloomington, IN 47405, USA. Fax: (1)
(812) 855-4520.
139 induced ascorbate release in the neostriatum is completely abolished 33. The crus cerebri, however, contain descending fibers from several forebrain structures, including cerebral cortex, making it difficult to specify the exact source of ascorbate release. In the present report, we combined voltammetric techniques with neurotoxic lesions of selective neuronal pathways to shed further light on the systems underlying the release of neostriatal ascorbate produced by amphetamine and other dopamine agonists. One group of animals received unilateral injections of 6-hydroxydopamine (6-OHDA) into the medial forebrain bundle (MFB) to destroy all ascending dopaminergic systems, including those projecting to the neostriatum. Separate animals sustained large-scale destruction of the neostriatum via unilateral infusions of kainate. Our results indicate that mechanisms intrinsic to the neostriatum, rather than brainstem dopaminergic projections, regulate the increase in neostriatal ascorbate produced by both direct and indirect dopamine agonists. MATERIALS AND METHODS
Animals Adult, male, Sprague-Dawley rats (350-450 g), bred in our animal colony, were used as subjects. They were housed individually under standard laboratory conditions with food and water available continuously.
Kainate and 6.0HDA lesions Prior to surgery, all animals were anesthetized with Equithesin (0.33 ml/kg, i.p.); diazepam (10 mg/kg; i.p.) was administered to kainate animals to minimize seizures while 6-OHDA animals were pretreated with the monoamine oxidase inhibitor, pargyline (40 mg/kgl i.p.), and the norepinephrine re-uptake blocker, desipramine (25 mg/kg; i.p.). The animals then were placed in a stereotaxic instrument and a hole was drilled unilaterally through the skull over the neostriatum (1.0 mm anterior and 2.5 mm lateral to bregmat~). For kainate lesions, a cannula was lowered into the neostriatum (5.0 mm ventral to the dura) and 1.0/zl (2.0 g g / t t l ) kainate was infused over a 5-rain period. For 6-OHDA lesions, a second hole was drilled over the MFB (4.0 mm posterior and 2.0 mm lateral to bregma), a cannula was lowered through this hole into the MFB (8 mm ventral to the dura) and 2.0/zl (4.0 tt~;//zl) 6-OIIDA was infused over a 5-rain period. After the infusion (kainate or 6-OHDA), the cannula was left in place for 5 rain. The cannula then was removed and a plastic cylindrical hub, designed to mate with an electrode holder for voltammetric recordings, was centered over the hole located above the neostriatum and cemented onto the skull. Prior to cementing, the hole over the MFB was filled with Gelfoam, an absorbable gelatin sponge. Control animals received the same treatment as lesion animals, including lowering the cannula into the brain, but no neurotoxin was administered. Following surgery, all animals were treated with 0.20 ml (25 rag) Rocephin (Roche) to minimize infection. Voltammetric recordings were performed after a recovery period of approximately 7 days.
Voltammetry Microvoltammetric electrodes were prepared from Thorneli P-55 carbon fibers (Union Carbide) as described previously7'12. The active
electrode surface is 200 ~ m in length with a 5-/~m radius. Each electrode was treated in a citrate-phosphate buffer solution by application of a triangular wave potential (70 Hz, 0.09-3.0 V vs. a saturated calomel reference electrode, SCE) for 20 s followed by 10 s of 0.5 V cathodal current, 3 s of 0.9 V anodal current, and 8 s of 1.5 V cathodal current. This treatment modifies the active surface of the electrode to allow discrimination of ascorbate from catechols and other easily oxidizable compounds 1°. Each electrode then was tested with staircase voltammetry in 100/zM ascorbate and 20/zM 3,4-dihydroxyphenylacetic acid (DOPAC) until the response stabilized. For in vivo voltammetry, the electrode was placed into a holder consisting of a stainless steel, luer-lock fitting that mates with the hub placed on the animal's skull during surgery. The holder is equipped with a threaded assembly that allows the electrode to be raised and lowered through the brain. For the experiments in this report, the electrode was lowered 5 mm ventral to the brain surface into the neostriatum. Generation of waveforms for staircase voltammetry and storage of sampled current was performed by a computer (IBM-XT) interfaced to a locally constructed potentiostat of conventional 3-electrode design6. The carbon-fiber electrode served as the working electrode, while the stainless steel electrode holder, which extends 2.5 mm ventral to the brain surface, served as both the reference and auxiliary electrodes. A potential was applied to the working electrode in 2.44-mV steps from 0 to 500 mV vs reference to insure the oxidation of ascorbate and DOPAC. Short pulses (98 ms) were used in order to restrict sampling to the solution pool near the electrode surface. The scan rate was set at 10 mV/s; scans were obtained at 2-min intervals. Voitammograms displayed peaks representing ascorbate (approximately 150 mV vs. reference) and DOPAC (approximately 300 mV vs. reference), as well as a third more positive peak thought to represent 5-hydroxyindolacetic acid and uric acid 4.5. Drug-induced changes in extracellular ascorbate were tested after a stable baseline was established and all animals were resting quietly. Separate groups of both lesion and control animals received either 2.5 mg/kg D-amphetamine (s.c.), 20.0 mg/kg GBR-12909 (s.c.) or the combination of 10.0 mg/kg SKF-38393 (i.p.) and 1.0 mg/kg quinpirole (s.c.). These doses of dopamine agonists are known to produce reliable increases in neostriatal ascorbate as well as characteristic patterns of locomotion and stereotyped behavior (see ref. 19). In all cases, voltammetric signals were recorded for another 50 min. When voltammetric recording was completed, all animals were anesthetized with sodium pentobarbital. Some of the working electrodes used for each drug group were removed for postcalibration in 100 /zM ascorbate and 20 ~M DOPAC. In order to verify the recording site, current was passed through the electrodes that remained in the brain in order to produce a small lesion. Following transcardial perfusion with normal saline and 10% formalin, brains were removed from the lesioned animals. Brain tissue was subsequently frozen, sectioned, and stained with Cresyl violet. For histological analysis, sections were observed microscopically (Carl Zeiss microscope; 16× and 40x objectives) for electrode-placement lesions and, when appropriate, extent of damage caused by kainate infusions. The extent of kainate lesions was determined by extent of gliosis and cell death in the neostriatum and surrounding structures. An acceptable kainate lesion was operationally defined as gliosis and cell death in at least 75% of the neostriatum with little or no damage outside of this structure (with the exception of cortical damage dorsal to the neostriatum, see Results).
Drugs D-Amphetamine sulfate (Sigma), quinpiroie (RBI), pargyline (Sigma), and kainate (Sigma) were mixed in 0.9% NaCI. GBR-12909 (RBI), SKF-38393 (RBI) and desipramine (Sigma) were mixed in 5% tartaric acid. 6-OHDA (Sigma) was mixed in a 0.9% NaCI-0.2% ascorbate solution. Diazepam (Sigma) was mixed in a vehicle containing 50% saline, 40% propylene glycol and 10% ethanol. Amphetamine and 6-OHDA were mixed as the free base, all other compounds were expressed as the salt.
140 Statistics
TABLE I
Voltammograms were expressed as percent baseline following drug treatment. Between-subjects analyses of variance were performed on all our data. Further pairwise comparisons were performed with Tukey's HSD.
Vohammetrically recorded extraceihdar levels (i~M +_S.E.M) of basal ascorbate and catechols in the neostriatum of control, kainate- and 6-OHDA-lesioned rats
The asterisk represents a significant difference from the control and kainate groups (P < 0.05, Tukey's HSD).
RESULTS All identifying lesions were located in anterior neostriatum: 4.0 to 5.0 mm ventral to the brain surface and approximately 2.5 mm lateral to the midline. As shown in Table I, based on postcalibration values, baseline extracellular neostriatal ascorbate levels were similar across treatment groups. These values are comparable with neostriatal ascorbate levels reported in previous studies3.10,27,32. Kainate a n d 6 - O H D A lesion verification
The effectiveness of kainate and 6 - O H D A lesions were verified histologically and biochemically, respectively. Fig. 1 depicts the typical extent of gliosis and cell death following a unilateral kainate lesion. In all cases, approximately 75% of the neostriatum was damaged. Table I includes voltammetric recordings not only of the ascorbate peak but also the catechol peak, which represents both dopamine and its metabolite, DOPAC. Animals in the control and kainate lesioned
Treatment
n
Ascorbate
Catechols
Control Kainate 6-OHDA
10 6 6
314 (24) 238 (49) 306 (38)
24.1 (2.2) 22.5 (3.0) 2.3 (1.5) *
groups had similar catechol oxidation peaks; these values are comparable with those reported in previous studies g'27. As expected, basal catechol levels decreased dramatically following destruction of neostriatal dopaminergic afferents via 6-OHDA. Indeed, the catechol values in the 6 - O H D A group represent a reduction of more than 95% relative to controls. Dopamine agonist-induced behaviors in kainate- and 6-OHDA- as well as sham-lesioned animals also were noted. The sham-lesioned animals displayed behaviors typically observed following administration of amphetamine and GBR-12909 (locomotion, sniffing, rearing and head bobbing) as well as the combined administration of SKF-38393 and quinpirole (sniffing, head
Fig. I. Photomicrograph of a coronal section (approximately 1.0 mm anterior to bregma) through the neostriatum of a kainate-lesioned rat brain showing typical extent of damage. The darkly stained areas (left side of the section) represent the cell death and gliosis resulting from the lesion. Note damage to approximately 75% of the neostriatum. Although some collateral damage occurred in the cortex dor-sal to the neostriatum, cortical damage alone had no effect on ascorbate release (see Results).
141 bobbing, biting and licking). Similar to previous resuits 2°, the onset of these behaviors was within 10 rain of drug administration and persisted for the entire observation period (50 min). The lesioned animals displayed behaviors similar to the sham-lesioned animals with the addition of circling, which was particularly intense in the 5-OHDA groups. This is not surprising since dopamine agonist-induced circling ipsilateral to a 6-OHDA-lesion is associated with unilateral striatal dopamine depletions in excess of 950~ 14'37. Similarly, unilateral striatal kainate lesions are known t6 produce circling contralateral to the lesioned side a°. While sham-lesioned SKF-38393-quinpirole animals displayed little locomotion, 6-OHDA- and kainate-lesioned animals circled to a similar extent regardless of the dopamine agonist administered (data not shown).
Effects of kainate and 6-OHDA lesions on dopamine agonist-induced ascorbate release Statistical analyses revealed no significant difference in extracellular ascorbate levels between both sham-lesioned groups 50 min after administration of either a m p h e t a m i n e ( t 2 = 0.83, n.s.) or vehicle (t 2 = 0.6, n.s.). In both cases, amphetamine caused at least a 50% increase in ascorbate, whereas ascorbate declined slightly following vehicle treatment. Thus, the respective amphetamine and vehicle responses of each shamlesioned group was combined for comparison with both kainate- and 6-OHDA-lesioned animals. The ControlAmph group in Fig. 2 shows the time course of the neostriatal ascorbate response following administration of 2.5 mg/kg D-amphetamine. This gradual increase in
0~0 &mA
ControI-Amph 60HDA-Amph
C l u E ] Koinote-Amph O m O Control-Vehicle
80. bJ u3 +1 q~ ,JC
40.
"~
20.
rn
O.
~
-20.
fj,,}
j6
60.
13- -40
I
I
I
I
I
10
20
30
40
50
Minutes P o s t - I n j e c t i o n
Fig. 2. Effects of 2.5 mg/kg D-amphetamine on ascorbate release in the neostriatum of kainate- (Kainate-Amph), 6-OHDA- (6-OHDAAmph) and sham-lesioned (ControI-Amph) animals as well as the effects of vehicle treatment in sham-lesioned (Control-Vehicle) animals. All responses are presented as the mean (+_S.E.) percent predrug baseline for each 10-min period following amphetamine or vehicle administration. Note the steady increase in extracellular ascorbate in both the ControI-Amph and 6-OHDA-Amph groups. Conversely, ascorbate in the Kainate-Amph group did not differ from the Control-Vehicle group.
Koinote .-.
80
~
40
~
20
"g
o
~
GBR
60HDA
~
Control
SKF-Quin
Fig. 3. Effects of direct (a combination of 10.0 mg/kg SKF-38393 and 1.0 mg/kg quinpirole) and indirect (20.0 mg/kg GBR-12909) dopamine agonists on neostriatai ascorbate in kainate, 6-OHDA, and control (sham-lesioned) animals. All responses are displayed as the mean (_+ S.E.) percent predrug baseline for the 6-min period 44-59 min following drug administration. Note the similarities between the responses of these agonists to that of amphetamine (see Fig. 2). Thus, whereas both direct and indirect dopamine agonists induce neostriatal ascorbate release in control and 6-OHDA animals, the kainate group shows a decrease from baseline. The * represents a significant difference from the kainate group.
ascorbate over the first 50 min of the drug response is typically observed following the administration of both indirect (amphetamine and GBR-12909) and direct (SKF-38393-quinpirole combination) dopamine agonists (see ref. 19). Also shown in Fig. 2 are the time courses for the effects of amphetamine in kainate- and 6-OHDA-lesioned animals, rhe overall ANOVA and subsequent pairwise comparisons showed no significant difference between kainate-lesioned and ControlVehicle animals. By contrast, there was a significant increase in ascorbate in the 6-OHDA-lesioned animals following amphetamine. Indeed, pairwise comparisons revealed that the ascorbate response of the 6-OHDA group was not significantly different from the ControlAmph group at any of the time points examined. Similar to the control groups outlined above, there were no significant differences between the sham kainate- and sham 6-OHDA-iesioned groups when administered either GBR-12909 ( t 2 = 2.26, n.s.) or the SKF.38393-quinpirole combination (t: = 0.68, n.s.) during voltammetric recordings. Therefore, these groups were combined into two control groups, respectively, for comparison to data obtained from lesioned animals. Results similar to those seen with amphetamine were observed following the administration of GBR-12909 and the SKF-38393-quinpirole combination, as shown in Fig. 3. Thus, while these agonists clearly increased neostriatal ascorbate 50 min after their administration in the Control and 6-OHDA groups, they had no effect on neostriatal ascorbate in kainate-lesioned animals. In addition to extensive neostriatal damage in the kainate group, there also was some damage to the
142 cortex dorsal to the neostriatum (see Fig. 1). To eliminate cortical damage as a factor underlying the inability of amphetamine to increase neostriatal ascorbate levels in this group, we inflicted discrete cortical lesions similar to those observed following neostriatal lesions in a separate group of animals (0.5 /.tg//.tl kainate unilaterally over 5 min; 1.0 mm anterior and 2.5 mm lateral to bregma, 2.0 mm ventral to dura). Voltammetric recordings 7 days following surgery revealed an increase (n = 2) in ascorbate similar to that observed in control animals (40% mean increase) 50 min after administration of 2.5 mg/kg o-amphetamine.
DISCUSSION Consistent with previous results, our data indicate that amphetamine ~'~L23 as well as other dopamine agonists ~9 increase behavioral activation as well as the level of ascorbate in the extracellular fluid of the neostriatum. Although the functional significance of this increase is not well understood, dopaminergic neurons appear to play little or no role in it. Thus, as reported for rats with 6-OHDA lesions of the neostriatum 6'~2, we found that the amphetamine-induced increase in neostriatal ascorbate release is not significantly attenuated by large-scale depletion of ncostriatal dopamine. Interestingly, however, we found, in agreement with previous evidence ~°, that dopamine depletions delayed somewhat the increase in ascorbate induced by amphetamine (see Fig. 2). Our results with 6-OHDA lesions also applied to OBR-12909 and to the SKF-38393-quinpirole combination, indicating that neither indirect nor direct dopamine agonists require an intact dopaminergic projection to the neostriatum in order to induce ascorbate release. Our results also confirm that ascorbate is not released from neostriatai dopaminergic terminals 6.~°a2. In fact, electrical stimulation of the MFB, which activates dopaminergic neurons ~a, does not induce neostriatal ascorbate release ~3'2s. Furthermore, neostriatal areas of high dopaminergic activity are inversely related to areas of high extracellular ascorbate 3. We used 6-OHDA lesions of the MFB, which are known to cause dramatic decreases in dopamine levels throughout the entire forebrain, not just the neostriaturn 3~''37. Thus, it is difficult to argue that residual dopamine in nucleus accumbens or other dopaminerich forebrain regions contributes to our results with either amphetamine or GBR-12909. Unless these drugs exert a postsynaptic dopamine-mimetic effect like SKF-38393 and quinpirole, which seems unlikely, amphetamine and GBR-12909 must act on non-dopamin-
ergic systems to promote ascorbate release in the neostriatum. Ample evidence appears to link ascorbate release with glutamate-containing neurons. Brain regions with high concentrations of glutamate, such as neostriatum, hippocampus, and cortex, also contain high levels of ascorbate s'29. Furthermore, glutamate infusions into each of these areas promotes ascorbate release saT. In addition, cortical lesions, which remove a major source of glutaminergic input to the neostriatum, significantly lower basal levels of ascorbate 2'!7. Collectively, these results indicate that ascorbate is released from glutaminergic axon terminals, but the mechanism by which both direct and indirect dopamine agonists influence this process remains to be established. Several lines of evidence suggest that neostriatal ascorbate release from corticostriatal terminals may be regulated by the substantia nigra, which influences cortical activity via relays in the thalamus. It is known, for example, that unilateral infusions of amphetamine into the substantia nigra cause a bilateral increase in neostriatal ascorbate release 33'34. Furthermore, similar to cortical lesions, bilateral lesions of the ventromedial thalamus, a main target of outputs from substantia nigra pars reticulata (SNR), lower both basal levels of ascorbate and the amount of ascorbate released by amphetamine in the neostriatum 2. Because amphetamine increases the firing rate of SNR neurons 25'26, this drug may alter thalamic and eventually cortical activity and in that way influences neostriatal ascorbate release. Activity in the SNR is strongly influenced by feedback neurons originating in the neostriatum 22. This pathway is largely destroyed by neostriatal kainate infusions, which also prevent ascorbate release by either indirect or direct dopamine agonists. These results are consistent with previous evidence that electrolytic lesions of the crus cerebri, which also destroy the neostriatonigral pathway, abolish the ability of amphetamine infused directly into the SNR to increasc neostriatal ascorbate release 33. Kainate lesions of the neostriatum, however, also destroy cholinergic interneurons, which may play a role in ascorbate release. For example, the cholinergic agonist, pilocarpinc, produces a dose-dependent increase in neostriatal ascorbate release, and this effect is reversed by scopolamine 15. In summary, our results support a growing body of evidence that ascorbate is not released from dopaminergic axon terminals nor are such neurons critical for the release of neostriatal ascorbate induced by either indirect or direct dopamine agonists. Rather, this process appears to be controlled by the neostriatum itself,
143 either via cholinergic interneurons or via~ projection neurons that modulate a circuit involving nigral, thalamic, and cortical neurons. Acknowledgements. This research was supported by NSF Grant BNS 91-12055. We also acknowledge the technical assistance of Paul Langley and the secretarial work of Faye Caylor. We also thank Dr. Dale Sengelaub for assistance with histological analysis.
REFERENCES 1 Adams, R.N., In t'it,o electrochemical measurements in the CNS, Prog. Neurobiol., 35 (1990) 297-311. 2 Basse-Tomusk, A. and Rebec, G.V., Corticostriatal and thalamic regulation of amphetamine-induced ascorbate release in the neostriatum, Pharmacol. Biochem. Behav., 35 (1990)55-60. 3 Basse-Tomusk, A. and Rebec, G.V., Regional distribution of ascorbate and 3,4-dihydroxyphenylacetic acid (DOPAC) in rat neostriatum, Brain Res., 538 (1991) 29-35. 4 Cespuglio, R., Faradji, H., Ponchon, J.-L., Riou, F., Buda, M., Gonon, F., Pujol, J.-F. and Jouvet, M., In vivo measurement by differential pulse voltammetry of extracellular 5-hydroxTindoleacetic acid in the rat brain, J. Physiol. 77 (1981) 327-332. 5 Crespi, F., Sharp, T., Maidement, N.T. and Marsden, C,A., Differential pulse voltammetry: simultaneous in vivo measurement of ascorbic acid, catechols and 5-hydroxyindoles in the rat striatum, Brain Res., 332 (1984) 135-138. 6 Ewing, A.G., Wightman, R.M. and Dayton, M.A., In vivo voltammetry with electrodes that discriminate between dopamine and ascorbate, Brain Res., 249 (1982) 361-370. 7 Ewing, A.G., Dayton, M.A. and Wightman, R.M., Pulse voltammetry with microvoltammetric electrodes, Anal. Chem., 53 (1981) 1842-1847. 8 Ghasemzedah, B., Cammack, J. and Adams, R.N., Dynamic changes in extracellular fluid ascorbic acid monitored by in vivo electrochemistry, Brain Res., 547 (1991) 162-166. 9 Gonon, F., Buda, M., Cespuglio, R., Jouvet, M. and Pujol, J.F., In vivo electrochemical detection of catechols in the neostriatum of anaesthetized rats: dopamine or DOPAC?, Nature, 286 (1980) 902-904. 10 Gonon, F., Buda, M., Cespuglio, R., Jouvet, M. and Pujol, J.F., Voltammetry in the striatum of chronic freely moving rats: detection of catechols and ascorbic acid, Brain Res., 223 (1981) 69-80. 11 Justice Jr., J.B., Introduction to in vivo voltammetry, in J.B. Justice Jr. (Ed.), Voltammetry in the Neurosciences, Principles, Methods, and Applications, Humana Press, Clifton, New Jersey, 1987, pp. 3-101. 12 Kamata, K., Wilson, R.L., AIIoway, K.D. and Rebec, G.V., Multiple amphetamine injections reduce the release of ascorbic acid in the neostriatum of the rat, Brain Res., 362 (1986) 331-338. 13 Kuhr, W., Wightman, R.M. and Rebec, G.V., Dopaminergic neurons: simultaneous measurements of dopamine release and single-unit activity during stimulation of the medial forebrain bundle, Brain Res., 418 (1987) 122-128. 14 Miller, R. and Beninger, R.J., On the interpretation of asymmetries of posture and locomotion produced with dopamine agonists in animals with unilateral depletions of striatal dopamine, Prog. Neurobiol., 36 (1991) 229-256. 15 Mueller, K., Voltammetric evidence in vivo of cholinergic modulation of extracellular ascorbic acid and uric acid in rat striatum, Brain Res., 408 (1987) 313-316. 16 Oh, C., Gardiner, T.W. and Rebec, G.V., Blockade of both DIand D2-dopamine receptors inhibits amphetamine-induced ascotbate release in neostriatum, Brain Res., 480 (1989) 184-189. 17 O'Neill, R.D., Fillenz, M., Sundstrom, L. and Rawlins, J.N.P., Voltammetrically monitored brain ascorbate as an index of excitatory amino acid release in the unrestrained rat, Neurosci. Lett., 52 (1984) 227-233.
18 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982. 19 Pierce, R.C. and Rebec, G.V., Stimulation of both DI and D2 receptors increases behavioral activation and ascorbate release in the neostriatum of freely moving rats, Eur. J. Pharmacol., 191 (1990) 295-302. 20 Pierce, R.C. and Rebec, G.V., Dopamine-, NMDA-, and sigmareceptor antagonists exert differential effects on basal and amphetamine-induced changes in neostriatai ascorbate and DOPAC in awake, behaving rats, Brain Res., in press. 21 Pierce, R.C., Rowlett, J.K., Bardo, M.T. and Rebec, G.V., Chronic ascorbate potentiates the effects of chronic haloperidol on behavioral supersensitivity but not D 2 dopamine receptor binding, Neuroscience, 45 (1991) 373-378. 22 Rebec, G.V., Electrophysiological pharmacology of amphetamine. In J. Marwah, (Ed.), Neurobiology of Drug Abuse, Monographs in Neural Sciences, Vol. 13, S. Karger, Basel, 1987, pp. 1-33. 23 Rebec, G.V., Changes in brain and behavior produced by amphetamine: a perspective based on microdialysis, voltammetry, and single-unit electrophysiology in freely moving animals. In R.R. Watson (Ed.), Biochemistry and Physiology of Substance Abuse, Vol. III, CRC Press, Boca Raton, FL, 1991, pp. 93-115. 24 Rebec, G.V., Centore, J.M., White, L.K. and Alloway, K.D., Ascorbic acid and the behavioral response to haloperidol: implications for the action of antipsychotic drugs, Science, 227 (1985) 438-440. 25 Rebec, G.V. and Groves, P.M., Apparent feedback from the caudate nucleus to the substantia nigra following amphetamine administration, Neuropharmacology, 14 (1975) 275-282. 26 Ruffieux, A. and Schultz, W., Dopaminergic activation of reticulata neurones in the substantia nigra, Nature, 285 (1980) 240-241. 27 Schenk, J.O., Miller, E., Gaddis, R. and Adams, R.N., Homeostatic control of ascorbate concentration in CNS extracellular fluid, Brain Res., 253 (1982) 353-356. 28 Stamford, J.A., Kruk, Z.L. and Millar, J., Ascorbic acid does not modulate stimulated dopamine release: in vivo voltammetric data in the rat, Neurosci. Lett., 60 (1985) 357-362. 29 Stamford, J.A., Kruk, Z.L. and Millar, J., Regional differences in extraceiiular ascorbic acid in the rat brain determined by high speed cyclic voltammetry, Brain Res., 299 (1984) 289-295. 30 Vecsei, L. and Beal, M.F., Comparative behavioral and neurochemical studies with striatal kainic acid- or quinolinic acid-lesioned rats, Pharmacol. Biochem. Behav., .~9 (1991)473-478. 31 White, L.K., Maurer, M., Kraft, M.E., Oh, C. and Rebec, G.V., lntrastriatal infusions of ascorbate antagonize the behavioral response to amphetamine, Pharmacol. Biochem. Behac., 36 (1990) 485-489. 32 Wightman, R.M., Brown, D.S., Kuhr, W.G. and Wilson, R.L., Molecular specificity of in vivo electrochemical measurements. In J. Justice (Ed.), Voltammetry in the Neurosciences, Humana Press, Clifton, NJ, 1987, pp. 103-138. 33 Wilson, R.L., Kamata, K., Bigelow, J.C., Rebec, G.V. and Wightman, R.M., Crus cerebri lesions abolish amphetamine-induced ascorbate release in the rat neostriatum, Brain Res., 370 (1986) 393-396. 34 Wilson, R.L and Wightman, R.M., Systemic and nigral application of amphetamine both cause an increase in extracellular concentration of ascorbate in the caudate nucleus of the rat, Brain Res., 339 (1985) 219-226. 35 Yount, S.E., Kraft, M.E., Pierce, R.C., Langley, P.E. and Rebec, G.V., Acute and long-term treatments alter extracellular ascorbate in neostriatum but not nucleus accumbens of freely moving rats, Life Sci., 49 (1991) 1237-1244. 36 Zahm, D.S., Compartments in rat dorsal and ventral striatum revealed following injection of 6-hydroxydopamine into the ventral mesencephalon, Brain Res., 552 (1991) 164-169. 37 Zigmond, M.J., Abercrombie, E.D., Berger, T.W., Grace, A.A. and Stricker, E.M., Compensation after lesions of central dopaminergic neurons: some clinical and basic implications, Trends Neurosci., 13 (1990) 290-295.
