Ž .Brain Research 792 1998 126–132

Research report

Differential effects of NMDA and non-NMDA antagonists on the activity ofaromatic L-amino acid decarboxylase activity in the nigrostriatal dopamine

pathway of the rat

Andrew Fisher, Christopher S. Biggs, Michael S. Starr )

Department of Pharmacology, The School of Pharmacy, 29-39 Brunswick Square, London WC1N 1AX, UK

Accepted 27 January 1998

Abstract

This study examined the acute effects of a variety of NMDA and non-NMDA antagonists on the activity of aromatic L-amino acidŽ . Ž . Ž .decarboxylase AADC in the corpus striatum CS and substantia nigra SN of the rat. Sixty min pretreatment with the high affinity

Ž . Ž .NMDA receptor-channel blockers MK 801 0.01, 0.1 and 1 mgrkg and phencyclidine 4 mgrkg elevated AADC activity in both theŽ . ŽCS and SN 2- to 3-fold . Even more striking increases in AADC were noted with 40 mgrkg amantadine 3.8-fold for CS, 9.0-fold for

. Ž . ŽSN , 40 mgrkg memantine 3.4-fold for CS, 3.1-fold for SN; 20 mgrkg no effect and 40 mgrkg dextromethorphan 3.4-fold for CS,. Ž . Ž .6.2-fold for SN, in 6r10 ‘responders’ . Similarly pronounced increases in AADC activity in CS 1.9-fold and SN 2.8-fold were

Ž . Ž . Ž .detected after administering clonidine 2 mgrkg . R-HA 966 5 mgrkg, not 1 mgrkg modestly raised AADC activity in CS 0.45-foldŽ . Žand not SN. Other drugs had no effect on the activity of the decarboxylase enzyme, including CGP 40116 1 and 5 mgrg , eliprodil 10

. Ž . Ž .mgrkg , NBQX 10 mgrkg, 30 min pretreatment and atropine 1 mgrkg . These experiments indicate that blocking the NMDAŽ .receptor-channel and to a lesser extent the glycine site or stimulating a -adrenoceptors, profoundly increases AADC activity, more2

especially in the SN than CS. By contrast, inhibiting the NMDA glutamate recognition or polyamine sites, AMPA or muscarinic receptorsis without effect on AADC in either brain region. The ability of amantadine and memantine to potentiate the antiparkinsonian actions ofL-DOPA in the clinic, may be due to facilitated decarboxylation of L-DOPA by the brain. q 1998 Elsevier Science B.V.

Keywords: Aromatic L-amino acid decarboxylase; NMDA antagonists; Clonidine; Atropine

1. Introduction

Glutamatergic neurones within the basal ganglia arewidely believed to exert an opposite effect on motorbehaviour to those of the dopaminergic nigrostriatal sys-tem. Thus, application of glutamate receptor agonists to

w xthe striatum depresses motility 17,33 , whereas glutamatereceptor antagonists can have the opposite effect and elicitan increase in locomotion not unlike that produced by

w xdopamine agonists 1,26 . The newly discovered motorstimulant properties of glutamate antagonists have arousedconsiderable interest, for it was quickly realised that suchdrugs might offer a viable alternative to dopaminergicagents in the therapeutic management of Parkinson’s dis-

w xease 11,19,34,35 .

) Corresponding author. Fax: q44-171-753-5902; E-mail:mstarr@cua.ulsop.ac.uk

Ž .Attempts to unravel the mechanism s by which gluta-Ž .mate antagonists, especially N-methyl-D-aspartate NMDA

antagonists, modulate motor behaviour, have uncovered acomplex functional interdependence of the glutamate and

w xdopamine systems of the basal ganglia 34,35 . Studiesperformed with the prototypical NMDA receptor-channelblocker MK 801, for instance, have shown that MK 801’srobust motor stimulant action is severely compromised in

w xdopamine-depleted animals 8,34,35 . This is probably be-cause, in intact animals, MK 801 increases dopamine cell

w x w xfiring 41 , as well as increasing the release 40 andw xturnover of dopamine 20 .

Although MK 801 is largely devoid of behaviouralactivity in parkinsonian animals, it nevertheless retains afunctional link with the dopamine system. This is evi-denced by the fact that MK 801 potentiates the antiparkin-sonian activity of L-DOPA in dopamine-depleted rodents

w xand primates 35 . How MK 801 does this is not clear. Atthe postsynaptic level MK 801 has been shown to potenti-

0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.Ž .PII S0006-8993 98 00129-2

( )A. Fisher et al.rBrain Research 792 1998 126–132 127

ate dopamine D1 and to suppress dopamine D2 receptor-w xmediated behaviours 34 . However, we have found that

the dose of MK 801 required to potentiate L-DOPA-in-duced motor stimulation in reserpine-treated mice, was anorder of magnitude lower than that needed to accentuate

w xD1-dependent motor responding 15 . These findings sug-gested to us that while an increase in postsynaptic D1receptor function by MK 801 may contribute to its poten-tiation of L-DOPA-induced locomotion, it is not likely tobe the only or the main effect of NMDA receptor block-ade. This conclusion is borne out by our recent observationthat a 150 nM concentration of MK 801, applied by adialysis probe to the substantia nigra of reserpine-treatedrats, greatly potentiated the concentration of dopamine thatcould be recovered in response to the concomitant perfu-sion of a threshold amount of L-DOPA through the dialysis

w xprobe 2 . Since this low concentration of MK 801 did notliberate dopamine by itself, we concluded that MK 801was probably acting by increasing the bioconversion ofL-DOPA to dopamine, via the enzyme aromatic L-amino

Ž .acid decarboxylase AADC . Thus, glutamate antagonistsmight be useful as adjuncts rather than as alternatives toL-DOPA therapy, by working presynaptically to increasethe rate of dopamine synthesis.

In the past few years, it has emerged that AADC isw xsubject to regulation by both dopamine 42–44 and gluta-

w x w xmate receptors 12 . Hadjiconstantinou et al. 12 reportedthat MK 801 enhanced AADC activity acutely within thenigrostriatal, but not the mesolimbic dopamine system.This result could therefore explain the ability of MK 801to potentiate L-DOPA in behavioural and biochemical ex-periments. These authors have not indicated, however,whether stimulation of AADC occurs with other glutamateantagonists, or whether it is unique to MK 801. Thepurpose of the present study, therefore, was to compare avariety of NMDA and non-NMDA antagonists, to learnmore about their acute effects on AADC activity in thesubstantia nigra and corpus striatum of the rat.

2. Materials and methods

2.1. Materials

Amantadine, atropine, clonidine, pargyline and phen-Ž .cyclidine were obtained from Sigma, UK. q -5-methyl-

Ž .10,11-dihydro-5H-dibenzo a,d -cyclohepten-5,10-imineŽ .MK 801 and dextromethorphan were supplied by Re-

Ž .search Biochemicals, Natick, while RS -3-amino-1-hy-Ž .droxypyrrolidin-2-one HA 966 was supplied by Tocris

ŽCookson, Bristol. Eliprodil Synthelabo Recherche, Bag-´. Ž .neux , R-DL- E -2-amino-5-phosphophonopentanoic acid

Ž .CGP 40116; Ciba Geigy, Basel , 2,3-dihydroxy-6-nitro-Ž . Ž7-sulphamoyl-benzo f -quinoxaline-dione NBQX; Novo

. Ž .Nordisk, Malov and memantine Merz, Frankfurt wereall obtained as gifts. All drugs were administered dissolved

in physiological saline. The solution of eliprodil was aidedŽ .with dimethylsulphoxide Sigma, UK , and that of NBQX

with a minimum quantity of 0.1 M sodium hydroxideŽ .Sigma, UK . All reagents used for the AADC assay wereof analytical grade and supplied by Sigma, UK. Reagents

Ž .used for high performance liquid chromatography HPLCwere also of analytical grade and supplied by Fluka,Germany.

2.2. Animals and drug treatment

Ž .Male Wistar rats A.R. Tuck, UK , weighing 240–260g were used for these experiments. The animals wereinitially housed in groups of six at 22"18C, under fluo-rescent lighting from 07.00–17.00 h, and allowed water

Ž .and rat chow ad libitum. Drugs or saline controls wereinjected i.p. in a volume of 1 mlrkg and the animalsreturned to their home cage, before being sacrificed by

Ž .guillotine 1 h later 30 min for NBQX . The dosing andtiming of drugs used in this study was based on previously

w xpublished behavioural data 34–36 . Some drugs causedpronounced motor stimulation to animals in the homecage, which was noted by the experimenter as eitherpresent or absent, but was not quantified further. Allexperiments were conducted in accordance with the Ani-

Ž .mals Scientific Procedures Act 1986.

2.3. Aromatic L-amino acid decarboxylase assay

Ž .Aromatic L-amino acid decarboxylase AADC was as-sayed by a modification of the method described by Na-

w xgatsu et al. 21 . The assay is based on the enzymicconversion of L-DOPA to dopamine which is then esti-mated by HPLC and electrochemical detection. Followingdecapitation brains were rapidly removed into ice-coldsaline and the substantia nigra and corpus striatum dis-sected out. The tissues were homogenized in 0.25 M

Ž .sucrose solution 1 ml for nigra, 3 ml for striatum , and thehomogenates centrifuged in Eppendorf tubes at 3000 r.p.m.for 10 min. One hundred ml of the supernatant wereassayed for protein by the Bradford modification of the

w xLowry method 4 , while 20 ml were assayed for AADCŽactivity as follows. An incubation mixture total volume

.400 ml was formed of the following: sodium phosphateŽ .buffer pH 7.2 ; L-DOPA, 0.5 mM; ascorbic acid, 0.1 mM;

pyridoxal-5X-phosphate, 0.01 mM; 2-mercaptoethanol, 1mM; EDTA, 0.1 mM; pargyline, 0.1 mM; enzyme, 20 ml.Blank values were determined by substituting water for theL-DOPA substrate. Incubation was carried out for 20 minat 378C and was stopped by the addition of 80 ml ice-cold0.5 M perchloric acid. The mixture was then centrifuged at3000 r.p.m. for 10 min and the resulting supernatant takenfor HPLC assay, or stored at y808C for assay at a laterdate. Full details of the HPLC assay have been given

w xelsewhere 2 . Enzyme activity was expressed as nmoldopaminermg proteinrh.

( )A. Fisher et al.rBrain Research 792 1998 126–132128

2.4. Data analysis

AADC activities in vehicle- and drug-treated rats wereŽ .compared by analysis of variance ANOVA followed by

Student’s t-test.

3. Results

3.1. Effects of drugs on AADC actiÕity

Basal levels of AADC activity were found to be broadlyŽsimilar for the substantia nigra SN; 66.0"5.7 nmol

. Ždopaminermg proteinrh and corpus striatum CS; 59.1."3.0; Fig. 1 in vehicle-treated control rats. MK 801

Ž .0.01–1 mgrkg stimulated AADC in a dose-dependent,bell-shaped fashion; the lowest dose was without effect,while a peak effect occurred at 0.1 mgrkg and tailed off at

Ž .1 mgrkg Fig. 1 . The increases in AADC were propor-tionately higher in the SN than in the CS. Phencyclidine, at4 mgrkg, raised AADC activities by roughly the same

Ž . Ž .amounts in both SN q156% and CS q171% .By far the greatest effects on AADC, however, were

obtained with 40 mgrkg amantadine, more especially inŽ . Ž .the SN q801% than in the CS q281%; Fig. 2 . Me-

mantine, 20 mgrkg, was ineffective in both regions. At 40mgrkg, memantine was similar in potency to amantadine

Ž .with respect to the CS q233% , but much less effectiveŽ .than amantadine in the SN q210% . The results for

Ž .dextromethorphan 40 mgrkg appeared to fall into twoŽdiscrete groups, which we have termed ‘responders’ 6r10

. Ž .rats and ‘non-responders’ 4r10 rats; Fig. 2 . The non-re-sponders, as the name implies, showed no evidence of anyalteration in AADC at either end of the nigrostriataldopamine axis, whereas the responders exhibited marked

Fig. 1. Potentiation of AADC activity in striatum and nigra by highaffinity NMDA receptor-channel antagonists. Each result is the mean"

S.E.M. of at least 6 determinations. Units of AADC activity are nmoldopaminermg proteinrh. PCPsphencyclidine. ))) P -0.001 vs. con-trols by ANOVA and Student’s t-test.

Fig. 2. Potentiation of AADC activity in striatum and nigra by lowaffinity NMDA receptor-channel antagonists. Each result is the mean"

S.E.M. of at least 6 experiments. Units of AADC activity are nmoldopaminermg proteinrh. Amant s amantadine; dextro s

Ž .dextromethorphan; Rs responders; NRsnon-responders ns4 only .))) P -0.001 vs. controls by ANOVA and Student’s t-test.

Ž .increases in AADC, more especially in the SN q515%Ž .than in the CS q227% .

Pretreatment with the NMDA polyamine site antagonistŽ .eliprodil 10 mgrkg , the competitive NMDA glutamate

Ž .site antagonist CGP 40116 1 and 5 mgrkg , or with theŽ .selective AMPA receptor antagonist NBQX 10 mgrkg ,

failed to alter decarboxylase activity in either brain regionŽ .Fig. 3 . R-HA 966, an antagonist of the strychnine-insen-sitive NMDA glycine site, induced a modest increase inAADC activity at 5 mgrkg and not 1 mgrkg, and then

Ž . Ž .only in the CS q46% and not SN Fig. 4 . By contrast,Ž .the a -adrenoceptor agonist clonidine 2 mgrkg markedly2Ž .increased AADC activity in the SN q181% , and to a

Ž . Ž .smaller extent in the CS q87% . Atropine 1 mgrkgŽ .was not found to modify AADC in either region Fig. 4 .

Fig. 3. Lack of effect on AADC activity of competitive and polyaminesite NMDA antagonists, or of AMPA receptor antagonism. Each result isthe mean"S.E.M. of at least 6 determinations. Units of AADC activityare nmol dopaminermg proteinrh. Elipseliprodil.

( )A. Fisher et al.rBrain Research 792 1998 126–132 129

Fig. 4. Mixed effects of NMDA glycine site antagonism, a -adrenoceptor2

agonism and muscarinic receptor antagonism on AADC activity in stria-tum and nigra. Each result is the mean"S.E.M. of at least 6 determina-tions. Units of AADC activity are nmol dopaminermg proteinrh. Clonsclonidine; Atropsatropine. ))) P -0.001 vs. controls by ANOVAand Student’s t-test.

3.2. Effects of drugs on behaÕiour

A marked behavioural arousal was noticed after admin-Ž .istration of MK 801 middle and highest dose , phencycli-

dine, amantadine and memantine, but not with any of theother agents tested.

4. Discussion

4.1. Prominent facilitation of AADC actiÕity by low affinityNMDA receptor-channel antagonists

The present study confirms and extends the preliminaryw xobservation of Hadjiconstantinou et al. 12 , that acute

blockade of central NMDA receptors with MK 801 ampli-fies the rate of decarboxylation of L-DOPA in the CS. Infact, our finding of a 57% increase in striatal AADC with1 mgrkg MK 801, closely matches the 46% increase inenzyme activity reported by these authors using the sameexperimental protocol. It would appear, however, that thisdose of MK 801 is not optimal, since a 10-fold lower doseof MK 801 was clearly much more effective, particularlywith regard to AADC activation in the SN, where anincrease of 114% was recorded. Although motor be-haviours were not specifically quantified in the presentstudy, it was apparent that rats treated with 0.1 mgrkgMK 801 became highly active, while those receiving 1mgrkg MK 801 exhibited severe ataxia and motor uncoor-dination. At behaviourally-relevant doses, therefore, wecan say that motor stimulation produced with the highaffinity NMDA receptor-channel blocker MK 801, is ac-companied by a substantial rise in the dopamine-synthesis-ing capability of the nigrostriatal dopamine system. Asimilar behavioural arousal and activation of AADC werealso noted with phencyclidine, 4 mgrkg, reinforcing the

suggestion that dopamine synthesis is tonically suppressedw xby glutamatergic neurones 9 , and is disinhibited in the

presence of a use-dependent NMDA receptor-channelblocker.

Of particular interest, however, was the observation thatlow affinity inhibitors of the NMDA receptor-channelcomplex, such as a amantadine and memantine, elicitedextraordinary rises in AADC activity. For instance, withamantadine the magnitude of the increases in AADC activ-ity approximated to q281% in CS and q802% in SN.Earlier investigations have only disclosed relatively minorrises in striatal AADC activity of 50–70%, following

Žtreatment with dopamine receptor antagonists e.g.,.Ž .sulpiride, SCH 23390 42–44 , which makes our data for

amantadine appear all the more remarkable. If this potentaction of amantadine is related to its ability to occludeNMDA receptors, then we must assume that under normalphysiological conditions dopamine synthesis only proceedsat a fraction of its maximum possible rate, because thetonic inhibition of dopamine synthesis by glutamatergicneurones in the basal ganglia is a robust one. It haspreviously been suggested that endogenous glutamate maylimit the activity of tyrosine hydroxylase by maintaining

w xthe enzyme in the dephosphorylated form 9 . A similarmechanism may operate with respect to glutamatergic

w xinhibition of AADC 44 .Memantine has a lower affinity for the NMDA recep-

tor-channel complex than MK 801, but has a 19.4-foldw xgreater affinity than amantadine 18 . In the present study,

memantine was less effective than amantadine but moreeffective than MK 801 at enhancing AADC, suggesting thelatter is directly related to the kinetics of binding to theNMDA receptor complex. The superior clinical profile ofamantadine is attributed to its low NMDA receptor affinityw x24,27 , which, as we have now found, is also commensu-rate with the largest potentiation of dopamine synthesis.

Dextromethorphan has a complex and interesting cen-tral pharmacology, including an ability to selectively blockNMDA and not non-NMDA-induced neuronal currentsw x Ž22,39 , as well as a high affinity for s receptors like

.phencyclidine . In this study, 40 mgrkg dextromethorphanclearly stimulated AADC activity in some rats and notothers. Our justification for treating these groups sepa-rately for statistical purposes, was that we had earlierobserved a differential sensitivity of mice to the morphinan

w xin behavioural experiments 16 . This could possibly bedue to inter-individual differences in the ability to convertdextromethorphan into the more active metabolite, dextror-

w xphan 25 . In our ‘responders’ dextromethorphan-inducedincreases in striatal and nigral AADC activities were pro-

Ž .found q238% and q515%, respectively and were com-parable to those obtained with the 1-aminoadamantanes.These increases, too, might reflect the rapid kinetics ofdextromethorphan binding to NMDA receptors, although a

Ž .participation of other receptors e.g., s receptors cannotbe ruled out. We would emphasise that dextromethorphan

( )A. Fisher et al.rBrain Research 792 1998 126–132130

did not stimulate ambulatory activity in this study, indicat-ing that a rise in AADC activity per se does not necessar-ily manifest as a change in the animal’s behaviour.

Although a substantial enhancement of AADC activityoccurred with all of the NMDA receptor-channel antago-nists that we tested here, other types of glutamate antago-nist had little or no effect on AADC, in spite of selectingdoses of the drugs that we knew to be centrally active in

w xbehavioural experiments 5,36 . For example, blocking theglutamate recognition site on the NMDA receptor withCGP 40116 failed to alter AADC in either brain region.Eliprodil was similarly inactive and so, too, was NBQX,indicating that neither inhibiting the NMDA polyaminesite nor the AMPA receptor influenced L-DOPA decar-boxylation. Glycine is known to be an obligatory coagonistwith glutamate for the NMDA receptor, but blocking thissite by administering the partial agonist R-HA 966 resultedin a barely significant increase in AADC activity in the CSand not SN. Given that the most effective agents used inthis study were all use-dependent NMDA antagonists, andblocked the associated open ion channel, could indicatethat the glutamatergic pathways they blocked were toni-cally active. Alternatively, the results could reflect differ-ences in the NMDA receptor pharmacology of the drugsconcerned, andror the heterogeneity in subunit make-upand distribution of NMDA receptors in the basal gangliaw x27 .

4.2. Effects of clonidine on AADC actiÕity

Clonidine was included in our study, since it has beenshown to interact synergistically with NMDA antagonistsw x w x37 and atropine 7 to relieve akinesia in reserpine-treated

Ž .mice. As it happened, the increases in CS q88% and SNŽ .q180% AADC produced by 2 mgrkg clonidine wereonly marginally less than those elicited by 0.1 mgrkg MK

Ž .801 q98% and q230%, respectively . How clonidinemodulates AADC activity in the brain is unclear, espe-cially since clonidine has been shown to suppress thelight-induced formation of new AADC enzyme in the

w xretina 31 . In the CS, a -adrenoceptor stimulation by2w xclonidine has been reported to inhibit acetylcholine 29 ,

w xbut not glutamate release 14 , which argues against cloni-dine interfering with glutamate neurotransmission at stri-atal synapses. On the other hand, clonidine is apparentlyable to suppress glutamate output in other parts of thebrain, and so this effect may have contributed to its

w xpotentiation of AADC activity in the SN 14 .In the CS, endogenous glutamate activates acetylcholine

w xrelease via NMDA receptors 10 , while acetylcholine inturn activates dopamine release via muscarinic receptorsw x13 . This raised the question of whether acetylcholinecould act as an intermediary in the glutamatergic control ofAADC activity, and whether clonidine might mimick MK801 by blocking acetylcholine release. Pretreatment withthe antimuscarinic agent atropine, however, failed to alter

the rate of dopamine formation from L-DOPA, mitigatingagainst a cholinergic link in the regulation of striatalAADC.

4.3. Functional significance of AADC increase in substan-tia nigra

The susceptibility of L-DOPA decarboxylation in theSN to modulation by glutamate antagonists has not previ-

w xously been reported, although Hadjiconstantinou et al. 12did find that MK 801 elevated the levels of mRNA fortyrosine hydroxylase and AADC in the ‘midbrain’, sug-gesting the glutamate antagonist had stimulated the synthe-sis of new enzyme. The present study now shows that themagnitude of decarboxylase activation by MK 801, aman-

Žtadine, dextromethorphan and clonidine but not phencycli-.dine or memantine , is considerably greater within the

dopamine cell body region than within the dopaminergicterminal fields.

The function of the small dopamine pool in dendriteshas tended to be overshadowed by the much larger quanti-ties of dopamine that exist in the axon terminals. Neurosci-entists are beginning to suspect, however, that this den-dritic dopamine may actually be more important as far asthe beneficial antiparkinsonian effects of L-DOPA are con-cerned. This hypothesis is illustrated by the recent micro-

w xdialysis data of Orosz and Bennett 23 , who remarked thatthe rotational response to systemically injected L-DOPA inunilaterally 6-hydroxydopamine-lesioned rats, was tempo-rally congruent with dopamine release in the lesioned SNrather than the lesioned CS. These and other authors haveconsequently concluded that the bioconversion of L-DOPAto dopamine, especially in the dopamine-deficient SN, maybe of greater importance than the striatum for the mainte-

w xnance of motor behaviour 23,30,32 . Our finding that thedecarboxylase capacity of the SN is open to such impres-sive modulation by glutamate antagonism, brings this per-spective of nigral dopamine into sharper focus.

4.4. Implications for the treatment of Parkinson’s disease

The possibility of using glutamate antagonists to relievethe motor deficiencies of Parkinson’s disease, and thepotential problems associated with this approach, have

Žbeen the subject of numerous reviews see Refs.w x.11,19,34,35 . If basal ganglia glutamatergic pathways areindeed hyperactive in parkinsonism, thereby causing targetneurones to become hyperexcitable, then monotherapy withglutamate antagonists might reasonably be expected torestore the glutamate balance to normal and to alleviate theassociated motor symptoms. In practice we find that anysuch beneficial action is generally masked by unwanted

w xside effects 34,35 . Nevertheless, the 1-aminoadamantanesŽ . w xand to a lesser extent dextromethorphan 3 , might wellowe their antiparkinsonian activity to just such a mecha-

w xnism 18 .

( )A. Fisher et al.rBrain Research 792 1998 126–132 131

Consequently, the most effective use of glutamate an-tagonists, both in the laboratory and the clinic, would

w xappear to be as adjuncts to L-DOPA 35 . The clinicallyuseful drugs amantadine and memantine are known toenhance the antiparkinson efficacy of L-DOPA in manw x28 , but the reason for this potentiation has never beensatisfactorily explained. Our present results now suggestincreased rate of decarboxylation of L-DOPA as a possiblemechanism, by increasing V andror the synthesis of themax

w xenzyme 43 . It must be remembered that in the parkinso-nian brain, the synthesis of dopamine from exogenouslyapplied L-DOPA is determined by the activity and amountof AADC in the neurones and glia, which decline for a

Ž .number of reasons: a continuing degeneration of theŽ . w x Ž .dopamine pathway; b loss of AADC with ageing 6 ; c

increased suppression of AADC activity by hyperactiveŽ .glutamatergic inputs; and d down-regulation of AADC

w xby long-term L-DOPA treatment 38 . These restrictionscould combine to make AADC activity rate-limiting in theparkinsonian brain, in which case the complementary useof a low affinity NMDA receptor-channel blocker, couldoffer a simple and effective way of boosting AADCactivity and extending the therapeutic lifetime of L-DOPA.

Acknowledgements

This work was supported by grants from The Sir JulesThorne Charitable Trust and the Medical Research Coun-

Ž .cil. Gifts of drugs from Ciba Geigy CGP 40116 , MerzŽ . Ž .memantine and Novo Nordisk NBQX are gratefullyacknowledged.

References

w x1 I. Al-Khatib, H.C. Karadag, A. Ulugol, The behavioural effects ofMK 801 injected into the nucleus accumbens and caudate-putamen

Ž .of rats, Pharmacol. Biochem. Behav. 52 1995 723–730.w x2 C.S. Biggs, L.J. Fowler, P.S. Whitton, M.S. Starr, NMDA receptor

antagonists increase the release of dopamine in the substantia nigraŽ .of reserpine-treated rats, Eur. J. Pharmacol. 299 1996 83–91.

w x3 U. Bonuccelli, P. Del Dotto, P. Piccini, F. Behge, G.U. Corsini, A.´Ž .Muratorio, Dextromethorphan and parkinsonism, Lancet 340 1992

53.w x4 M.M. Bradford, A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dyeŽ .binding, Anal. Biochem. 72 1976 248–254.

w x5 S. Brooks, S. Kaur, B.S. Starr, M.S. Starr, Motor actions of eliprodilin the normal and monoamine-depleted mouse: a role in the treat-

Ž .ment of Parkinson’s disease?, J. Neural Transm. 103 1996 737–748.w x6 R. Brus, Activities of some enzymes which synthesize and metabolise

catecholamines in the brain and peripheral organs in developing rats,Ž .Arch. Immunol. Ther. Exp. Wars. 22 1975 449–457.

w x7 M. Carlsson, A. Carlsson, Marked locomotor stimulation inmonoamine-depleted mice following treatment with atropine in com-

w x Ž .bination with clonidine, J. Neural Transm. P-D Sect. 1 1989317–322.

w x8 B.V. Clineschmidt, G.E. Martin, P.R. Bunting, N.L. Papp, CentralŽ .sympathomimetic activity of q -5-methyl-10,11-dihydro-5H-di-

w x Ž .benzo a,d cyclohepten-5,10-imine MK 801 , a substance with po-tent anticonvulsant, central sympathomimtic, and apparent anxiolytic

Ž .properties, Drug Dev. Res. 2 1982 135–145.w x9 J.-M. Desce, G. Godeheu, T. Galli, J. Glowinski, A. Cheramy,´

Opposite presynaptic regulations by glutamate through NMDA re-ceptors of dopamine synthesis and release in rat striatal synapto-

Ž .somes, Brain Res. 640 1994 205–214.w x10 M.G. Giovannini, F. Camilli, A. Mundula, L. Bianchi, M.A. Col-

ivicchi, G. Pepeu, Differential regulation by N-methyl-D-aspartateand non-N-methyl-D-aspartate receptors of acetylcholine release from

Ž .the rat striatum in vivo, Neuroscience 65 1995 409–415.w x11 J.T. Greenamyre, C.F. O’Brien, N-Methyl-D-aspartate antagonists in

Ž .the treatment of Parkinson’s disease, Arch. Neurol. 48 1991 977–981.

w x12 M. Hadjiconstantinou, Z.L. Rossetti, T.A. Wemlinger, N.H. Neff,Dizocilpine enhances striatal tyrosine hydroxylase and aromaticL-amino acid decarboxylase activity, Eur. J. Pharmacol. Mol. Phar-

Ž .macol. Sect. 289 1995 97–101.w x13 F. Hery, M.F. Giorguieff, M. Hamon, M.J. Besson, J. Glowinski,

Role of cholinergic receptors in the release of newly synthesizedamines from serotonergic and dopaminergic terminals, in: S. Garat-

Ž .tini, J.F. Pujol, R. Samanin Eds. , Interactions Between PutativeNeurotransmitters in the Brain, Raven Press, New York, 1978, pp.39–51.

w x14 Y. Kamisaki, T. Hamahashi, T. Hamada, K. Maeda, T. Itoh, Presy-naptic inhibition by clonidine of neurotransmitter amino acid release

Ž .in various brain regions, Eur. J. Pharmacol. 217 1992 57–63.w x15 S. Kaur, M.S. Starr, Role of D receptor mechanisms in the potentia-1

tion of motor responses to L-DOPA and apomorphine by MK 801 inw x Ž .the reserpine-treated mouse, J. Neural Transm. P-D Sect. 8 1994

107–116.w x16 S. Kaur, M.S. Starr, Antiparkinsonian action of dextromethorphan in

Ž .the reserpine-treated mouse, Eur. J. Pharmacol. 280 1995 159–166.w x17 T. Klockgether, L. Turski, Towards an understanding of the role of

glutamate in experimental parkinsonism: agonist-sensitive sites inŽ .the basal ganglia, Ann. Neurol. 34 1993 585–593.

w x18 J. Kornhuber, J. Bormann, M. Hubers, K. Rusche, P. Riederer,¨Effects of the 1-aminoadamantanes at the MK-801 binding site ofthe NMDA receptor gated ion channel: a human postmortem study,

Ž .Eur. J. Pharmacol. Mol. Pharmacol. Sect. 206 1991 297–300.w x19 K.W. Lange, P. Riederer, Glutamatergic drugs in Parkinson’s dis-

Ž .ease, Life Sci. 55 1994 2067–2075.w x20 W. Loscher, R. Annies, D. Honack, The N-methyl-D-aspartate recep-¨ ¨

tor antagonist MK 801 induces increases in dopamine and serotoninmetabolism in several brain regions of rats, Neurosci. Lett. 128Ž .1991 191–194.

w x21 T. Nagatsu, T. Yamamoto, T. Kato, A new and highly sensitivevoltammetric assay for AADC activity by high-performance liquid

Ž .chromatography, Anal. Biochem. 100 1979 160–165.w x22 R. Netzer, P. Pflimlin, G. Trube, Dextromethorphan blocks N-

methyl-D-aspartate-induced currents and voltage-operated inwardŽ .currents in cultured cortical neurons, Eur. J. Pharmacol. 238 1993

209–216.w x23 D. Orosz, J.P. Bennett, Simultaneous microdialysis in striatum and

substantia nigra suggests that the nigra is a major site of action ofL-dihydroxyphenylalanine in the ‘hemiparkinsonian’ rat, Exp. Neu-

Ž .rol. 115 1992 388–393.w x24 C.G. Parsons, G. Quack, I. Bresink, L. Baran, E. Przegalinski, W.

Kostowski, O. Krzascik, S. Hartmann, W. Danysz, Neuropharmacol-Ž .ogy 34 1995 1239–1258.

w x25 G. Pfaff, P. Briegel, I. Lamprecht, Interindividual variation in theŽ .metabolism of dextromethorphan, Int. J. Pharmacol. 38 1983 618–

623.w x26 R.C. Pierce, G.V. Rebec, Intraneostriatal administration of glutamate

antagonists increases behavioural activation and decreases neostri-

( )A. Fisher et al.rBrain Research 792 1998 126–132132

atal ascorbate via nondopaminergic mechanisms, J. Neurosci. 13Ž .1993 4272–4280.

w x27 R.H.P. Porter, J.T. Greenamyre, Regional variations in the phar-macology of NMDA receptor channel blockers: implications for

Ž .therapeutic potential, J. Neurochem. 64 1995 614–623.w x28 J.M. Rabey, P. Nissipeanu, A.D. Korczyn, Efficacy of memantine,

an NMDA receptor antagonist, in the treatment of Parkinson’sw x Ž .disease, J. Neural Transm. P-D Sect. 4 1992 277–282.

w x29 P. Rada, E. Pothos, G.P. Mark, B.G. Hoebel, Microdialysis evidencethat acetylcholine in the nucleus accumbens is involved in morphine

Ž .withdrawal and its treatment with clonidine, Brain Res. 561 1991354–356.

w x30 G.S. Roberston, H.A. Roberston, Evidence that L-DOPA-inducedrotational behaviour is dependent on both striatal and nigral mecha-

Ž .nisms, J. Neurosci. 9 1989 3326–3331.w x31 Z. Rossetti, D. Krajnc, N.H. Neff, M. Hadjiconstantinou, Modula-

tion of retinal aromatic L-amino acid decarboxylase via a adreno-2Ž .ceptors, J. Neurochem. 52 1989 647–652.

w x32 S. Sarre, P. Herregodts, D. Deleu, A. Devrieze, N. De Klippel, G.Ebinger, Y. Michotte, Biotransformation of L-DOPA in striatum andsubstantia nigra of rats with a unilateral, nigrostriatal lesion: amicrodialysis study, Naunyn-Schmiedebergs Arch. Pharmacol. 346Ž .1992 277–285.

w x33 W.J. Schmidt, D. Bury, Behavioural effects of N-methyl-D-aspartateŽ .in the anterodorsal striatum of the rat, Life Sci. 43 1988 545–549.

w x34 M.S. Starr, Glutamaterdopamine D rD balance in the basal gan-1 2Ž .glia and its relevance to Parkinson’s disease, Synapse 19 1995

264–293.w x35 M.S. Starr, Antiparkinsonian actions of glutamate antagonists—alone

and with L-DOPA: a review of evidence and suggestions for possiblew x Ž .mechanisms, J. Neural Transm. P-D Sect. 10 1995 141–185.

w x36 M.S. Starr, B.S. Starr, Comparison of the effects of NMDA andAMPA antagonists on the locomotor activity induced by selectiveD and D dopamine agonists in reserpine-treated mice, Psychophar-1 2

Ž .macology 114 1994 469–476.w x37 A. Svensson, M.L. Carlsson, Injection of the competitive NMDA

receptor antagonist AP-5 into the nucleus accumbens ofmonoamine-depleted mice induces pronounced locomotor stimula-

Ž .tion, Neuropharmacology 31 1992 513–518.w x38 H. Takashima, M. Tsujihata, M. Niwa, S. Nagataki, Effects of

chronic administration of L-DOPA and bromocriptine on thedopaminergic metabolism in the rat striatum, Biogenic Amines 4Ž .1987 15–22.

w x39 F.C. Tortella, M. Pellicano, N.G. Bowery, Dextromethorphan andneuromodulation: old drug coughs up new activities, Trends Phar-

Ž .macol. Sci. 10 1989 501–507.w x40 P.S. Whitton, Glutamatergic control over brain dopamine release in

Ž .vivo and in vitro, Neurosci. Biobehav. Rev. 21 1997 481–488.w x41 J. Zhang, L.A. Chiodo, A.S. Freeman, Electrophysiological effects

of MK-801 on rat nigrostriatal and mesoaccumbal dopaminergicŽ .neurons, Brain Res. 590 1992 153–163.

w x42 M.Y. Zhu, A.V. Juorio, I.A. Paterson, A.A. Boulton, Regulation ofaromatic L-amino acid decarboxylase by dopamine receptors in the

Ž .rat brain, J. Neurochem. 58 1992 636–641.w x43 M.-Y. Zhu, A.V. Juorio, A. Paterson, A.A. Boulton, Regulation of

striatal aromatic L-amino acid decarboxylase: effects of blockade orŽ .activation of dopamine receptors, Eur. J. Pharmacol. 238 1993

157–164.w x44 M.-Y. Zhu, A.V. Juorio, Aromatic L-amino acid decarboxylase:

biological characterization and functional role, Gen. Pharmacol. 26Ž .1995 681–696.