Movement Disorders
Vol. 19, No. 7, 2004, pp. 743–754
© 2004 Movement Disorder Society
Research Review
Recent Failures of New Potential Symptomatic Treatments for
Parkinson’s Disease: Causes and Solutions
Gurutz Linazasoro, MD*
Centro de Neurologı́a y Neurocirugı́a funcional, Clı́nica Quirón, San Sebastián, Guipúzcoa, Spain
Abstract: One major goal of current research in Parkinson’s
disease (PD) is the discovery of novel agents to improve
symptomatic management. The object of these new treatments
should be to provide effective symptom control throughout the
course of the disease without the development of side effects
such as motor and psychiatric complications. Results of several
clinical trials of new treatment options reported in the past 2
years have shown negative or unsatisfactory results. Most of
the drugs and surgical procedures used in these studies had
been tested previously in 1-methyl-4-phenyl-1,2,3,6-tetrahy-
dropyridine (MPTP) monkeys as well as in the classic 6-hy-
droxydopamine–lesioned rat model. They raise several ques-
tions about the true reliability of animal studies, the adequacy
A major goal of current research in Parkinson’s dis-
ease (PD) is the discovery of novel agents to improve
symptomatic management in the hope that new treat-
ments will provide effective and sustained symptom con-
trol throughout the course of the disease, without induc-
ing side effects such as motor and psychiatric
complications.1 To reach this objective, it is important to
have relevant models of the disease in which new phar-
macological agents and surgical procedures could be
tested before clinical assessment. In some ways, PD has
been considered paradigmatic for other pathological con-
ditions, particularly neurodegenerative illness, and has
the advantage of some unique features. For instance, the
*Correspondence to: Dr. Gurutz Linazasoro, Centro de Neurologı́a y
Neurocirugı́a funcional, Clı́nica Quirón, Parque Alcolea s/n, 20012 San
Sebastián, Guipúzcoa, Spain. E-mail: glinazasoro@terra.es
Received 14 September 2003; Revised 23 December 2003; Accepted
2 February 2004
Published online 21 April 2004 in Wiley InterScience (www.
interscience.wiley.com). DOI: 10.1002/mds.20120
743
of the working hypotheses and design of clinical trials, the
validity of tools in current use to evaluate a specific effect, and
the selectivity of the drugs used. All these factors may explain
failure. This review focuses on pharmacological and surgical
treatments tested to improve the management of patients with
motor fluctuations and dyskinesias. Some of the recent trials
and possible reasons for their lack of success are critically
analysed. Finally, some suggestions to avoid further failures
and improve results are proposed. © 2004 Movement Disorder
Society
Key words: Parkinson’s disease; motor complications; ani-
mal models; treatment alternatives
degenerative process is quite limited to the dopaminergic
neurons localised in the pars compacta of the substantia
nigra (SNpc), and damaging the nigrostriatal pathway at
different levels in animals reproduces the striatal dopa-
mine (L-dopa) deficiency characteristic of PD.2 Rodent
models were the first available and proved very useful as
early screens in the development of antiparkinsonian
therapy. The unilateral 6-hydroxydopamine (6-OHDA)-
lesioned rodent, through the ability to induce rotational
behaviour with apomorphine or amphetamine, has a high
predictive value to detect potential anti-parkinsonian
agents.3 Because rotational behaviour does not resemble
the motor abnormalities seen in PD, new methods have
been developed in recent years to better assess these and
also levodopa-induced dyskinesias.4,5
The possibility of inducing parkinsonism in subhuman
primates with the administration of 1-methyl-4-phenyl-
1,2,5,6-tetrahydropiridine (MPTP)6 led to the belief that
the predictability would be greatly improved because the
evaluation of motor, and to some extent nonmotor, be-
744 G. LINAZASORO

haviour would be closer to that of humans.7,8 The use-
fulness of these models has been reviewed recently.3,6
With the help of these models, further insights into the
pathophysiology of parkinsonism and dyskinesia have
significantly contributed to the formulation of more ac-
curate hypotheses as starting points to develop new treat-
ment approaches.3,6 Despite high hopes of discovering
more effective pharmacological and surgical approaches,
the results of several clinical trials reported in the past 2
years do not uphold such expectations. Indeed, most of
them have shown negative or at best unsatisfactory re-
sults when tried in PD patients (Table 1). Most of the
drugs and surgical procedures used in these studies had
been tested previously in MPTP monkeys as well as the
classic 6-OHDA–lesioned rat model. The results of all
these studies raise several questions about the true reli-
ability and validity of animal data, the adequacy of the
current working hypotheses, and the presently used tools
to evaluate a specific effect.
NEW PHARMACOLOGICAL TREATMENTS
FOR MOTOR FLUCTUATIONS AND
LEVODOPA-INDUCED DYSKINESIAS:
RECENT CLINICAL TRIALS
This article focuses on pharmacological and surgi-
cal treatments to improve the management of patients
already suffering from motor fluctuations and dyski-
nesia. New concepts on the organization of the basal
ganglia and the influence of different neurotransmit-
ters and neuromodulators on motor function9,10 have
led to the development of new drugs with potentially
beneficial effects against parkinsonism and L-dopa–
induced dyskinesias.
Dopaminergic Agents
Several new dopaminergic drugs are in advanced clin-
ical development and will soon be released to the market:
sumanirole, rotigotine, L-dopa methylester and ethyl-
ester, and the triple combination of L-dopa/carbidopa/
entacapone. Piribedil, which has been in use in France
for more than 2 decades, will become available in other
countries, and new drugs designed to improve the dopa-
minergic transmission to produce a better clinical re-
sponse are in the pipeline.
Drugs Acting at the Dopamine Transporter Level.
These drugs increase the synaptic levels of dopamine
by blocking the action of the synaptic dopamine trans-
porter.11 The higher the levels of dopamine in the syn-
apse, the longer and stronger the antiparkinsonian effect.
Brasofensine is one of these drugs, producing a potenti-
ation of the effects of L-dopa in MPTP-lesioned monkeys
without inducing dyskinesia.12 In 8 patients receiving the
drug, an initial improvement over a 4-week period was
observed, but this improvement has been lost by the end
of the study, and no significant differences were shown
against placebo. A possible explanation of this may be
that the metabolism of brasofensine in human beings is
different from that in monkeys, invalidating all preclin-
ical toxicological investigations.12 Other drugs with the
same mechanism of action, such as NS2330, are now
being developed.

TABLE 1. Treatment strategies with symptomatic potential in complicated PD:

Comparison between animal and human data
Animal data Human data
Park LID Park LID
Dopamine uptake blockers (brasofensine) ⫹⫹⫹ ⫹⫹⫹ - ??
D1 receptor agonists ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ -
Partial D2 agonists ⫹⫹⫹ ⫹⫹⫹ ⫹/⫺ ⫹/⫺
Adenosine A2a antagonists ⫹⫹⫹ ⫹⫹⫹ ⫹⫹ ⫹/⫺
Alfa 2 antagonists - ⫹⫹⫹ - ⫹/⫺
GABAergic ??? ??? ⫹/⫺ ⫹/⫺
Antiserotoninergic drugs (5HT2c) ⫹⫹⫹ ⫹⫹⫹ ?? ??
Serotoninergic drugs (5-HT1a) - ⫹⫹⫹ - ⫹⫹⫹
Glutamate antagonists ⫹⫹⫹ ⫹⫹⫹ ⫹/⫺ ⫹/⫺
Cannabinoid agonists ⫹⫹⫹ ⫹⫹⫹ - ⫹⫹
Cannabinoid antagonists - ⫹⫹⫹ ?? ??
Opioid antagonists - ⫹⫹⫹ - -
Opioid agonists ⫹⫹ - - -
Functional surgery ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹
Fetal transplantation ⫹⫹⫹ ⫹⫹⫹ ⫹/⫺ -

Park, parkinsonism; LID, levodopa-induced dyskinesias; ⫹⫹⫹, good effect; ⫹⫹, moder-
ate effect; ⫹/⫺, slight effect; -, no effect; PD, Parkinson’s disease.

Movement Disorders, Vol. 19, No. 7, 2004

 FAILURES OF SYMPTOMATIC TREATMENTS FOR PD
Selective Dopamine Receptor Agonists.
D1 receptors have been repeatedly involved in the
origin of L-dopa–induced dyskinesia,13,14 and selective
D1 receptors agonists have been shown to be highly
effective in MPTP monkeys, producing a natural motor
response without dyskinesia. Even more, D1 agonist did
not produce dyskinesias in parkinsonian monkeys al-
ready primed to develop dyskinesias,15 and this change
occurring after a short period of administration might be
interpreted as a reversal of the underlying mechanism.
Some confusion still exists about the real motor effect of
D1 agonists due to the previous use of drugs without a
full D1 profile, such as CY 20824316 and SKF 38393.
Dihydrexidine was the first high-affinity full D1 agonist
tested in both animals17 and humans,18 and it produced a
dramatic reduction of parkinsonian signs in severely
afflicted MPTP-treated monkeys17; in the only clinical
trial reported, 4 patients received the drug intravenous-
ly18: three of them suffered severe postural hypotension
and were withdrawn from the study; the remaining pa-
tient showed a 70% improvement in UPDRS score. The
significant drop in blood pressure precluded human stud-
ies.
ABT-431 (along with isochromans A77636 and
A86930), similar in structure and pharmacological prop-
erties to dihydrexidine and from which a comparable
effect might be expected, were the next D1 agonists
investigated. These agents had been used in MPTP mon-
keys with good effect against parkinsonism and dyski-
nesia.19 However, ABT-431, as with dihydrexidine, had
poor oral bioavailability, although it was tolerated better
than the latter. When tested in humans with PD and
motor complications, ABT-431 had a similar action to
L-dopa, but without significant modification of dyskine-
sia.20,21 These results were somewhat disappointing, be-
cause expectation had been high, particularly concerning
its potential antidyskinetic effect. Isochromans have not
been tested in PD patients; they do not have the problems
of bioavailability reported with dihydrexidine and ABT-
431 but can cause severe side effects and behavioural
tolerance. Dinapsoline (DAR-201) has good oral avail-
ability, does not induce tolerance, and has been shown to
be efficacious in rats and marmosets with parkinson-
ism.22
Partial Agonists.
This class of compounds that may act either as ago-
nists or antagonists, depending on the sensitivity of do-
pamine receptors: they exert agonist activity at dener-
vated supersensitive sites and competitively inhibit
complete agonists at fully innervated normosensitive re-
ceptors. Thus, their therapeutic window appears to be
745
higher than full agonists mainly in patients with L-dopa–
induced dyskinesia or psychiatric disturbances. Ter-
guride is an example of a partial D2 agonist exhibiting
both antiparkinsonian and antidyskinetic effects in
MPTP-treated monkeys,23 but when administered to pa-
tients with motor complications the results were incon-
sistent. In some cases, terguride reduced dyskinesia with-
out worsening parkinsonism, but in others, it worsened
both dyskinesia and PD.24
Miscellaneous.
Other drugs acting at the dopaminergic system
through varied mechanisms of action are being explored,
although none of them have as yet been tested in humans.
Among these drugs are dopamine stabilisers (such as
OSU-6162), which are drugs with a weak D2 receptor
antagonistic capacity, depending on the dopaminergic
tone,25 D3, D4, and D5 agonists, D3 antagonists, and so
on. Recent data indicate the existence of a linear increase
of the expression of D3 receptors in relation to the
severity of dyskinesia in MPTP monkeys receiving L-do-
pa.26 Dyskinesia was alleviated with BP 897, a partial
agonist at the D3 receptor, when, importantly, the ben-
eficial therapeutic response to L-dopa was not affected.26
In contrast, full D3 antagonists, either nonselective or
highly selective, also improved dyskinesia but at the
expense of a worsening of parkinsonism.26 While these
results support the use of D3 partial agonists in combi-
nation with L-dopa in patients with L-dopa–induced dys-
kinesia, other studies have shown that D3 antagonists
may exert antiparkinsonian effects by their own.27 Such
contradictions in animal models add confusion to the
expected effects when these drugs are eventually tested
on patients.
NONDOPAMINERGIC STRATEGIES
Adenosine A2a Receptor Antagonists
Adenosine A2a receptors are localised to striatal cho-
linergic interneurons and also to the cell bodies of striatal
␥-aminobutyric acid (GABA)ergic neurons, which orig-
inate the indirect pathway28; they are, therefore, colo-
calised with D2 receptors.28 Through this selective lo-
calisation, A2a receptors can influence both striatal
GABA and acetylcholine release.
Adenosine A2a agonists stimulate the indirect path-
way, increasing GABA release in the external globus
pallidus (GPe), whereas antagonists lead to a reduction in
the levels of GABA in the GPe. Overactivity of the
indirect pathway is one of the hallmarks of parkinson-
ism,9,10 so A2a antagonists could be beneficial in motor
symptoms. As these receptors are apparently absent in
Movement Disorders, Vol. 19, No. 7, 2004
746 G. LINAZASORO
the striatal neurons, which initiate the direct pathway and
in other dopaminergic pathways, these drugs could be
free of the major side effects of other antiparkinsonian
drugs.28 This hypothesis has been proved in 6-OHDA
rats and MPTP monkeys when the animals experienced a
more natural motor response than with L-dopa.28 KW-
6002, VER-11135, SCH-58261, and SCH-63390 are
some A2a antagonists tested that did not induce dyski-
nesia even in primed animals.
The expectations raised by these preclinical studies
have been only partially confirmed in two small clinical
trials with KW-6002 in humans.29,30 Eighty-three PD
patients with motor fluctuations and dyskinesia were
included in a double-blind randomized study with three
arms (placebo, 20 mg/d KW6002, and 40 mg/d
KW6002).29 The number of hours per day off were
slightly reduced (0.5–1.5 hours); dyskinesia remained
unchanged throughout the study. Nausea was the most
common side effect, and 7% of those participating ex-
perienced an increase in lipase level for no known rea-
son.29 In another study of 16 parkinsonian patients with
motor complications, a 40% potentiation of the L-dopa
effect was observed, with no increase in dyskinesia30; in
this study, L-dopa was administered as an infusion.
The case of theophylline is confusing: like caffeine, it
is not a specific adenosine antagonist. Open-label studies
have shown that it could ameliorate PD symptoms with-
out worsening dyskinesia.31 However, these results have
not been confirmed in a double blind study.32
Alfa2 Noradrenergic Receptor Antagonists
The role of noradrenaline and adrenaline in motor
function is far from clear: even the distribution and role
of adrenergic receptors within the basal ganglia nuclei is
not totally known. Despite this background of uncer-
tainty, it was found that idazoxan, a selective alfa2 re-
ceptor antagonist, extended the motor action of a dose of
L-dopa in MPTP-lesioned monkeys and reduced L-dopa–
induced dyskinesia.33 Similar effects have been reported
recently with JP-173034 and yohimbine,35 and it has been
suggested that these effects are mediated through an
action on the indirect pathway. Two clinical trials have
been conducted to date36,37: in the first one, the effect of
idazoxan on an apomorphine challenge was assessed in 8
patients with PD and motor complications.36 Four of
them experienced severe side effects, such as headache,
nausea, vomiting, and hypotension and did not complete
the study. In the remaining 4 patients, a nonsignificant
effect was found in both, duration of the motor benefit
and dyskinesias. In the other study, 18 patients with PD
were assessed by using a L-dopa challenge.37 In this
study, the on time did not change but a benefit against
Movement Disorders, Vol. 19, No. 7, 2004
dyskinesias was observed. Thus, results are contradictory
and several explanations can be offered: different actions
of L-dopa and apomorphine, and distinct neural mecha-
nisms underlying dyskinesia induced by L-dopa or apo-
morphine.38 A prospective multicentric study including a
large sample of fluctuating PD patients has been stopped
because of a lack of efficacy in an interim analysis. What
is clear from these small studies is that tolerability,
particularly regarding cardiovascular side effects, should
be improved, and animal studies heightened this expec-
tation.
Mirtazapine is a special case because it acts on the
noradrenergic and serotoninergic transmission. A few
observations in patients have suggested that it might
alleviate tremor and dyskinesia by a still unknown mech-
anism,39 but this benefit may be due to the antidepressant
and sedative effects of this drug.
GABAergic Drugs
The pathophysiological basis of motor complications
is still a matter for debate. It has been suggested that
chronic and pulsatile administration of L-dopa might
result in an imbalance between the direct and indirect
pathways connecting the striatum to the internal segment
of the globus pallidus (GPi), leading to the appearance of
motor fluctuations and dyskinesia.1 GABA is the main
neurotransmitter of both the indirect and direct striato-
pallidal pathways: numerous GABAergic neurons are
found in the human basal ganglia as well as GABAa and
GABAb receptors. GABAa receptors upregulation in the
GPi correlates with the development of dyskinesias in-
duced by L-dopa or dopamine agonists in monkeys40,41
and PD patients.42,43 Also, direct injection of muscimol
(a GABA agonist) on the thalamus, GPi, and subthalamic
nucleus (STN) in patients mimics the therapeutic effects
of lesions or deep brain stimulation (DBS) of these
nuclei.44 However, the influence of orally administered
GABAergic drugs on parkinsonian motor disability is
contradictory. While early morning dystonia can be im-
proved by baclofen,45 other GABAergic compounds
have shown no efficacy or have worsened PD symp-
toms.46 – 48 Gabapentin increases the synthesis and re-
lease of GABA and may induce motor side effects, such
as chorea and dystonia,49 suggesting that it may act on
the circuitry involved in motor control. We have shown
that gabapentin improves the basal off motor score,
thereby reducing the strength of the motor response after
an L-dopa dose50 but other clinical and pharmacological
parameters are not improved and the overall clinical
situation of patients remains unchanged. Previous studies
also suggested a moderate beneficial effect of gabapentin
in PD.51,52 To further complicate this scenario, it has
 FAILURES OF SYMPTOMATIC TREATMENTS FOR PD
been shown that flumazenil, a GABA antagonist, may
improve features of PD.53 This was an open study in
which 8 patients received an acute, intravenous dose of
flumazenil. These confusing results illustrate the diffi-
culty in predicting the consequences of the manipulation
of the GABA system. It is the main inhibitory neuro-
transmitter in the central nervous system and GABA
receptors are found in virtually all the basal ganglia
nuclei. Therefore, a presumed positive effect derived
from the action on GABA receptors located in a given
nucleus may be hampered by the action on the same
receptor located in a different nucleus. Consequently
until more specific GABA drugs are synthesised, no firm
conclusion about efficacy and mechanism of action can
be made.
Drugs Acting on Serotoninergic Transmission
The basal ganglia, mainly the striatum and the GPi/
substantia nigra pars reticulata (SNpr) complex, receive a
dense serotoninergic innervation mainly from the brain-
stem raphe nucleus. Among the different subtypes of
these receptors, 5HT2c is the most interesting for PD.54
This receptor is located at its highest concentration in the
GPi and SNpr, and it has been shown that its binding is
increased in postmortem brain tissue of parkinsonian
patients.54 Thus, targeting this receptor may be of benefit
in PD as can be inferred from the 5HT2c knockout rat,
which displays spontaneous hyperlocomotion55 and has
been also shown in lesioned rats.56 At present, no selec-
tive 5HT2c antagonists are marketed and no human data
are available, although some atypical neuroleptics with
this action, such as clozapine, are less prone to induce
parkinsonism. In line with this, actions on this same
receptor but located in the STN have been claimed to be
responsible for oral dyskinesias, which were reverted
with a selective antagonist drug.57 Ritanserin, a selective
5HT2 antagonist, significantly ameliorated L-dopa–in-
duced dyskinesias in a single-blind, placebo-controlled,
study performed in 10 patients,58 but this result has not
been replicated.
After Durif and colleagues59 noted that fluoxetine
could improve L-dopa–induced dyskinesia, other sub-
types of serotonin receptors (striatal 5HT1a receptors)
have been implicated in the appearance of dyskinesia.
However, this single study was short and open, and the
effect of fluoxetine could be related to its tendency of
worsening parkinsonism.60 In a recent open label study,
more than 400 parkinsonian patients with depression
were given sertraline and neither parkinsonism nor L-
dopa–induced dyskinesia were modified.61 Conceivably,
serotoninergic 5-HT1a autoreceptors, by regulating the
release of serotonin as well as dopamine formed from
747
exogenous L-dopa, may affect the response alterations
complicating L-dopa treatment of PD. Sarizotan is a
selective 5-HT1a agonist, which showed a potent anti-
dyskinetic effect without affecting parkinsonism in ani-
mal studies.62 Peak dose choreiform dyskinesias were
reduced by more than 90%.62 These results have been
replicated in 17 patients,63 but long-term studies are
needed.
Glutamate Antagonists
Glutamate is the main excitatory neurotransmitter in
the brain.64 – 66 Two major types of glutamate receptors
have been identified: ionotropic and metabotropic. Iono-
tropic are coupled to ion channels and can be divided in
N-methyl-D-aspartate (NMDA), AMPA, and kainate re-
ceptors. Metabotropic receptors are coupled to G pro-
teins. All these types of receptors are composed of mul-
tiple subtypes (NR1 and NR2 belong to the NMDA
family, mGluR1 to mGluR8 belong to the metabotropic
family, etc.). The precise location and function of many
of these subtypes are not totally known, most of them are
distributed in various nuclei of the basal ganglia and in
other brain structures, and there is a lack of drugs with an
adequate selectivity profile for a given receptor. Thus,
the resultant effects of any glutamate antagonist may be
absolutely different if administered locally in a nucleus
or systemically.64 – 66 In addition, this lack of precision
may produce severe side effects. This finding emphasises
the difficulties in attributing an effect to a glutamate
antagonist and explains the possibility of obtaining con-
tradictory results. Finally, glutamate has two faces: it is
essential to the normal development and function of the
brain, participating in key functions such as memory
formation and many others, but in certain circumstances,
it may be deleterious. But too little is as harmful as too
much, because it participates actively in the natural re-
covery process after a lesion as has been shown in
stroke.67 Caution is needed before using these drugs in
clinical trials.
There is a large amount of evidence supporting the
active role played by glutamate in the origin of parkin-
sonism and L-dopa–induced dyskinesia.68 For instance,
the changes in the phosphorylation state of glutamate
receptors as a result of the denervation and the pulsatile
treatment with L-dopa seem to be critical in the origin of
L-dopa–induced dyskinesia.68 As with other classes of
drugs, the antiparkinsonian and antidyskinetic profile of
this type of compounds has been demonstrated elegantly
in MPTP-treated monkeys,64 – 66 where these effects have
been shown with a large number of agents acting quite
selectively at NR2B,69 NR2A, AMPA, and mGluR570
receptor site. But many of these compounds cannot be
Movement Disorders, Vol. 19, No. 7, 2004
748 G. LINAZASORO
used in humans, because of side effects, or when tested,
have produced poor results. For instance, ifenprodil and
Ro 25-6981 behaved remarkably in MPTP monkeys, but
ifenprodil was not efficacious in PD patients.71 Remace-
mide was able to potentiate the effects of L-dopa in
animal models,72 but when used in PD patients already
taking L-dopa, no significant effect was seen in parkin-
sonism or dyskinesia.73 Riluzole is another NMDA an-
tagonist that has been given for motor complications
after showing good effect in animals,74 and a 30% re-
duction in dyskinesia in a small trial of 6 parkinsonian
patients,75 but a large, prospective multicenter trial has
been discontinued because lack of efficacy was detected
in an interim analysis. Dextromethorphan has produced
confusing results in humans, in part due to its narrow
therapeutic window,76 lamotrigine was ineffective in a
double-blind trial77 but amantadine has been shown to
ameliorate L-dopa–induced dyskinesia in acute and long-
term studies,78,79 an effect that has been attributed to its
antiglutamatergic properties, although amantadine is a
“dirty” drug with many other actions.
New antiglutamatergic agents are at a very early
stages of development, but they have to overcome all the
problems already mentioned before coming into clinical
use. To solve these problems, we have to increase our
knowledge at the molecular level about the involvement
of glutamate in the pathophysiology of parkinsonism and
dyskinesia, which would led to the development of se-
lective drugs with a positive risk/benefit profile.
Cannabinoids Agonists and Antagonists
The levels of cannabinoid receptors 1 (CB1) in the
basal ganglia are the highest in the brain, comparable to
dopamine receptors.80,81 This distribution of receptors is
consistent with their profound effects on motor function,
besides their well known involvement in other functions,
such as cognition and pain. Their main effect is to inhibit
the release of the neurotransmitter, by blocking the up-
take of GABA.80,81 The motor effects of cannabinoids
are complex.80 – 83 Cannabinoids activate movement
when microinjected into the striatum or into the SNpr,
and movement is inhibited when they are injected di-
rectly into the STN or the GPe. Recent studies in animal
models suggest that CB1 receptor antagonists could
prove useful in the treatment of parkinsonian symptoms
and L-dopa–induced dyskinesia and cannabinoid agonists
may be of great value as antidyskinetic agents.80,81,84
This suggestion has to be accepted with great caution
because the final effect of a systemically administered
CB1 agonist or antagonist will depend on the clinical
situation (not treated vs chronically treated patients),
Movement Disorders, Vol. 19, No. 7, 2004
which produces different compensatory modifications in
distinct nuclei.
We have shown that the systemic administration of the
CB1 agonist WIN55512 to parkinsonian rats produced a
decrease in the firing rate of STN neurons, which was
abolished by injection of the antagonist SR141716A.85
An opposite effect was observed in intact rats, a finding
also obtained with dopamine agonists.85,86 Moreover,
WIN55512 did not alter the neurophysiological charac-
teristics of nigral dopaminergic neurons. These data sug-
gest that CB1 agonists may exert antiparkinsonian ac-
tions, whereas antagonists may ameliorate L-dopa–
induced dyskinesia. By contrast, results of other
experiments81 and of the only clinical trial to date, failed
to support our data.87 In this study, nabilone, a cannabi-
noid agonist was given to 9 patients with L-dopa–induced
dyskinesia in a double-blind, randomized, placebo-con-
trolled, crossover study.87 Nabilone produced a mean
20% reduction in the dyskinesia score without modifying
parkinsonian symptoms and signs; it was particularly
effective in reducing diphasic dyskinesia and, to a lesser
extent, off period dyskinesia. A possible explanation of
these contradictory results is that the antidyskinetic ef-
fect of nabilone might be related to an action on the GPe
CB1 receptors.81,87 Again, the apparently consistent re-
sults obtained in animals could not be replicated in
humans; clearly further hypotheses need to be formu-
lated and ideally tested on animals before continuing
with clinical studies.
Opioid Receptor Agonists and Antagonists
Opioid peptides are involved in many aspects of basal
ganglia function, as can be inferred from the wide dis-
tribution of opioid receptors within these structures.
Changes in the expression of striatal opioids (enkephalin
and dynorphin) colocalised with dopamine receptors on
the GABAergic projection neurons occur in very early
phases of PD and further changes are induced by the
treatment with L-dopa.88 But problems mentioned for
other systems are applicable here: there are different
opioid receptor types located on different nuclei, of
which stimulation or blockade result in variable effects
on motor behaviour. Opioids act as neuromodulators
influencing the activity of dopamine and GABA neurons
in different directions, depending on the nucleus and
receptor involved (reviewed in Henry and Brotchie89).
Immunocytochemical and PET studies have demon-
strated increased opioid neuropeptide transmission in
animals and patients with PD and L-dopa–induced dys-
kinesia.88 –90 Thus, opioid receptor antagonists have been
studied as potential new treatment for dyskinesia.89 Mu
(cyprodime) and delta (naltrindole) selective receptor
 FAILURES OF SYMPTOMATIC TREATMENTS FOR PD
antagonists, have been shown to be efficacious in im-
proving dyskinesia in primed MPTP-lesioned primates.89
Very little effect has been observed in the few patients
who have been treated with naloxone91 and naltrexone.92
Enadoline, a kappa opioid agonist, has been demon-
strated to exert an anti-akinetic action and to potentiate
the motoric effects of L-dopa in reserpinized rats.93 To
date, the only data on human beings were obtained with
spiradoline, a kappa agonist with lower specificity and
potency than enadoline (see Hughes and colleagues93).
No antiparkinsonian effects were found in this study, but
the interpretation of the results were difficult because
dose-limiting behavioural abnormalities appeared. Thus,
as is the case with many of the drugs reviewed, a better
understanding of the function of the opioid system and
the synthesis of more selective compounds is needed.
SURGICAL TREATMENTS FOR PD PATIENTS
WITH MOTOR COMPLICATIONS
Pallidotomy and Deep Brain Stimulation
Current surgical procedures are based on the existence
of hyperactivity at the level of the subthalamopallidal
pathway,9,10 a feature that is considered to be the phys-
iological hallmark of the parkinsonian state. Abolition of
this hyperactivity by lesioning or inhibiting neural activ-
ity by implanting electrodes in the STN or the GPi results
in a dramatic improvement of parkinsonism in animal
models of the disease.9,10 Lesioning procedures, such as
pallidotomy and, to a lesser extent subthalamotomy, and
pallidal and subthalamic DBS have been used in parkin-
sonian patients,94 and as predicted, major improvements
in motor function have been obtained. However, several
aspects are not totally understood: (1) pallidal surgery
abolishes L-dopa–induced dyskinesias. This finding is
probably the most striking effect of any type of surgical
procedure into the GPi and is hard to reconcile with
previous models of the organization of the basal ganglia-
thalamo-cortical circuits in which only firing rate data
were considered.9 According to these models, any type
of dyskinesia is associated with a reduction in the activ-
ity of the subthalamopallidal pathway, and a further
reduction of this activity will lead to the appearance or
worsening of such dyskinesias, a finding that is not
compatible with surgical experience. New models have
incorporated changes in the pattern of neuronal activity
and neuronal synchronization as crucial features of the
organization of these circuits.95 Thus, abolition of abnor-
mal neuronal patterns and “normalisation” of existent
abnormal rhythms in these nuclei in PD patients with
dyskinesias can be the underlying basis of the improve-
ment in L-dopa–induced dyskinesia. In one sense, it is
749
preferable to live without basal ganglia than with abnor-
mally functioning basal ganglia (“No news is better than
bad news”). This is a clear example of adaptation to new
experimental data; further adaptations will be needed to
explain new findings. (2) Surgical techniques do not
produce cognitive adverse effects. This idea relies on the
well-demonstrated segregation of functions across the
basal ganglia-thalamo-cortical circuits.9,10,95 Usually, as-
sociative and limbic functions of nuclei conforming the
basal ganglia are related to the activity of neurons lo-
cated on their ventral part, which are connected with
cortical areas functionally related to cognitive and emo-
tional functions. Because lesions and electrodes are
placed on motor portions of these nuclei, only motor
effects are expected. And, in fact, the motor effects of
these procedures are dramatic.94 However, their impact
on cognitive and emotional functions are controversial:
whereas some authors have failed to find major effects,
others have found concerning adverse effects.96 Some
factors seem to influence the appearance of these adverse
effects, such as the age of patients, the existence of prior
subclinical cognitive disturbances, or abnormalities on
magnetic resonance imaging.97 Cognitive and emotional
functions have not been adequately addressed in the
larger studies conducted to date and the true frequency of
cognitive and emotional disturbances may be underesti-
mated.98
Transplantation
The success of cell replacement for the treatment of
PD is based on two hypotheses: first, the predominant
symptoms develop because of the dysfunction or loss of
the dopaminergic neurons in the nigrostriatal pathway;
second, dopaminergic neurons grafted into the dopam-
ine-deficient striatum can replace the neurons lost as a
result of the disease process and can reverse, at least in
part, the major symptoms of the disease.99 Extensive
animal studies have demonstrated that immature dopa-
mine neurons survive, re-establish dopaminergic inner-
vation when grafted into the denervated striatum, and
restore baseline dopamine synthesis and release.99 Re-
sults of a series of small open-label trials with carefully
selected patients indicated that human fetal dopaminer-
gic neurons can survive and function in the striatum of
patients with PD and most of these patients showed a
remarkable and long-lasting clinical improvement. Nev-
ertheless, two double-blind sham-surgery controlled tri-
als have failed to reproduce predicted results; there has
been only modest efficacy and a significant number of
patients have developed severe off-state dyskinesias after
transplantation.100,101 Paradoxically, experiments in
6-OHDA–lesioned rats with L-dopa–induced dyskinesia
Movement Disorders, Vol. 19, No. 7, 2004
750 G. LINAZASORO

have shown that grafting dopaminergic fetal cells into
the striatum produced an improvement in motor behav-
iour, including dyskinesia, which was accompanied by a
reversal of the molecular mechanisms theoretically re-
sponsible for dyskinesia.4 Once again, a substantial dis-
cordance between animal data and human reality is ev-
ident. For this reason, it would be interesting to perform
transplantation experiments in animals with dyskinesias
before clinical studies. Curiously, transplantation of
other type of cells such as human pigmentary epithelial
retinal cells102 or the direct striatal infusion of trophic
factors, such as glial-derived neurotrophic factor
(GDNF),103 have led to remarkable improvements in a
few PD patients. As was the case with fetal neural
transplantation, clinical benefit is accompanied by incre-
ments in the uptake of fluorodopa in PET studies. No off
period dyskinesias have been reported in these patients,
but many questions still remain about the exact mecha-
nism of action and long-term effects of these procedures,
and double-blind, sham-controlled studies with these
cells and GDNF are now planned.
SOME CLUES TO UNDERSTANDING
THERAPEUTIC FAILURES
What are the main reasons for these pitfalls and con-
tradictory results? Likely, this is a multifactorial prob-
lem. Both the 6-OHDA rat and the MPTP primate are
quite adequate models of the disease.3,6 In fact, all do-
paminomimetic drugs exert very stereotyped effects in
these models that can be easily reproduced and that
predict their later effectiveness. Nonetheless, it should be
recognized that these models are not perfect, and these
imperfections could explain some recent failures. More-
over, currently, we have an incomplete understanding of
basal ganglia functional organization on which to base
preclinical investigation of new treatments.9,10,94,95 Nev-
ertheless, it has produced some major advances in ther-
apy. For instance, the demonstration of the existence of
a neuronal hyperactivity in the STN and GPi/SNpr in
experimental models of PD led to the hypothesis that
reducing or abolishing this activity would result in an
improvement of parkinsonian cardinal symptoms and
signs. And this strategy was the origin of current DBS
techniques, which have changed the quality of life of
many patients with advanced PD. However, attempts to
test other hypotheses have failed. Hence, limitations of
the models, problems of current understanding of func-
tional organization of basal ganglia, and questions re-
lated to the pharmacological properties of the drugs
tested can explain recent pitfalls (see Table 2). For in-
stance, current animal models are usually models of an
acute dopaminergic deficit, although more chronic intox-
ication protocols have been developed.2,6,104 Of interest,
the effects of certain drugs are quite different in monkeys
with acute or chronic MPTP parkinsonism.105 However,
they are not totally comparable with PD, which is a
chronic, slowly progressive, and “more than dopaminer-
gic” disease. At present, there is no evidence to show that
chronic administration of low dose of any toxic sub-
stance mimics the progressive degenerative nature of PD.
Therefore, animal models offer a static rather than a
dynamic perspective of the disease. Moreover, the to-
pography of neuronal loss within the SNpc is different in
animals acutely lesioned or intoxicated from that in
PD,106,107 a difference less marked in chronic animal
models of PD.2,104 Finally, the clinical picture of PD
varies greatly from patient to patient, whereas the behav-
iour of parkinsonian rats or monkeys is very homoge-
neous.
The effects of L-dopa and other antiparkinsonian drugs
may be totally different, depending on the severity of
nigrostriatal denervation. Rodent and primate models of
moderate striatal dopaminergic deficiency are avail-

TABLE 2. Some problems with current methods

A) Related to animal models and assessment methods.
Animals are models of an acute, stable, severe, and pure dopaminergic deficiency.
Animals display a homogeneous behavioural disturbance.
Genetic factors, aging-related factors, etc., cannot be properly addressed in animals.
Cognitive, emotional and other nondopaminergic signs are difficult to evaluate in animals.
B) Related to limited knowledge of basal ganglia function.
Incomplete knowledge of the distribution and function of many receptors (and, perhaps, neurotransmitter systems) within the basal ganglia.
C) Related to treatment strategies.
Incomplete knowledge of the mechanism of action of certain drugs (i.e., levodopa).
Lack of drugs with complete selectivity for a given receptor in a given localization.
Action of drugs on receptors located outside the basal ganglia and involved in motor control or other functions.
Interactions between different neurotransmitters at the molecular level after the action of a drug.
Partial or dual effects of a drug are very difficult to evaluate.
Pharmacokinetic and pharmacodynamic characteristics of tested drugs.
Incomplete understanding of the mechanism of action of surgical therapies.

Movement Disorders, Vol. 19, No. 7, 2004

 FAILURES OF SYMPTOMATIC TREATMENTS FOR PD
able.2,6,86,104 Partial lesion protocols provide ideal mod-
els to evaluate and predict the effects of some therapies
on the prevention of the development of motor compli-
cations and dyskinesia. For instance, working with the
triple combination of L-dopa/carbidopa/entacapone, it
has been shown that q.i.d administration to marmosets
with severe dopamine deficiency produces a maximal
antiparkinsonian benefit with minimal dyskinesia com-
pared to L-dopa/carbidopa alone,108 but the difference is
not so evident when the same drug regimens are admin-
istered to marmosets with a moderate dopaminergic de-
nervation. We have also seen that the effects of L-dopa
and apomorphine on the neuronal activity of the STN is
more homogeneous in moderately 6-OHDA–lesioned
rats.86 There is also controversy about the earlier use of
DBS techniques or cell therapy approaches. So it would
be important to have data on animal with less-extensive
nigral lesions. The influence of other factors such as the
genetic background, the effects of aging, and so on,
cannot be adequately addressed in these models.
PD is a multiple system disorder and the degenerative
process is not limited to the dopaminergic nigrostriatal
pathway.109 Other cell populations containing different
neurotransmitters, such as the noradrenergic locus cer-
uleus,110 the serotoninergic raphe nucleus, the cholin-
ergic and glutamatergic pedunculopontine nuclei, and the
cholinergic nucleus of Meynert are also affected. This
pattern of involvement may account for the appearance
of some motor and nonmotor symptoms and signs. The
involvement of some of these nuclei may be very early,
earlier even than the SNpc damage.110 These degenera-
tive changes are not reproduced in current experimental
models.
Most of the drugs tested act at the level of neurochem-
ical systems, of which the relative contribution to the
pathophysiology of motor abnormalities has not been
totally elucidated. It is also unknown if there exists
always a human counterpart of a given receptor in a
given nucleus. For instance, the existence of dopamine
receptors into the human STN has not been conclusively
proven,111 the colocalization of D1 and D2 receptors in a
proportion of striatal neurons112 is ignored, etc.
All these arguments clearly limit the predictive value
of animal studies. With the increased complexity of
drugs acting on other neurotransmission systems, there is
a still lower predictive value as illustrated by the case of
fetal transplantation.100,101 Furthermore, it is important to
make further efforts to better define the function of the
basal ganglia in health and disease, along with the de-
velopment of alternative models of the disease.113,114 All
these efforts should be completed with the implementa-
tion of thoroughly conducted development programs.
751
To improve current symptomatic treatments with new
drugs, cell lines in which test the effects of drugs on the
molecular mechanisms underlying parkinsonism and
dyskinesia can be developed so that drugs can be tested
before conducting the “motor” experiments in ani-
mals.115 Finally, new technologies such as DNA mi-
croarrays,116 RNA silencing or RNA interference and
high-throughput screening of drugs may provide new
tools to develop more selective antiparkinsonian drugs
and to refine current models.
REFERENCES
1. Obeso JA, Linazasoro G, Gorospe A, et al. Complications asso-
ciated with chronic levodopa therapy in Parkinson’s disease. In:
Olanow CW, Obeso JA, editors. Beyond the decade of the brain.
Vol. 2. Kent: Wells Medical Ltd; 1997. p 11–36.
2. Tolwani RJ, Jakowec MW, Petzinger GM, et al. Experimental
models of Parkinson’s disease: insights from many models. Lab
Anim Sci 1999;49:363–371.
3. Cenci MA, Whishaw IQ, Schallert T. Animal models of neuro-
logical deficits: how relevant is the rat? Nat Rev Neurosci 2002;
3:574 –579.
4. Lee CS, Cenci MA, Sculzer M, Bjorklund A. Embryonic ventral
mesencephalic grafts improve levodopa-induced dyskinesia in a
rat model of Parkinson’s disease. Brain 2000;123:1365–1379.
5. Andersson M, Hilbertson A, Cenci MA. Striatal fosB expression
is causally linked with L-dopa-induced abnormal involuntary
movements and the associated upregulation of striatal prodynor-
phin mRNA in a rat model of Parkinson’s disease. Neurobiol Dis
1999;6:461– 474.
6. Jenner P. The contribution of the MPTP-treated primate model to
the development of new treatment strategies for Parkinson’s
disease. Parkinsonism and related disorders 2003;9:131–137.
7. Burns RS, Chiueh CC, Markey SP, et al. A primate model of
parkinsonism: selective destruction of dopaminergic neurons in
the pars compacta of the substantia nigra by N-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine. Proc Natl Acad Sci U S A 1983;80:
4546 – 4550.
8. Burns RS, Markey SP, Phillips JM, Chiueh CC. The neurotoxic-
ity of MPTP in the monkey and man. Can J Neurol Sci 1984;11:
166 –168.
9. DeLong MR. Primate models of movement disorders of basal
ganglia origin. Trends Neurosci 1990;13:281–285.
10. Brotchie JM. The neural mechanisms underlying levodopa-in-
duced dyskinesia. Ann Neurol 2000;47(Suppl. 1):105–114.
11. Jaber M, Jones S, Giros B, Caron MG. The dopamine transporter:
a crucial component regulating dopamine transmission. Mov Dis-
ord 1997;12:629 – 633.
12. Pearce RKB, Smith LA, Jackson MJ, et al. The monoamine
reuptake blocker brasofensine reverses akinesia without dyskine-
sia in MPTP-treated and levodopa-primed common marmosets.
Mov Disord 2002;5:877– 886.
13. Mailman R, Huang X, Nichols DE. Parkinson’s disease and D1
dopamine receptors. Curr Opin Invest Drugs 2001;2:1582–1591.
14. Boyce S, Rupniak NM, Steventon MJ, Iversen SD. Differential
effects of D1 and D2 agonists in MPTP-treated primates: func-
tional implications for Parkinson’s disease. Neurology 1990;40:
927–933.
15. Pearce RK, Jackson M, Britton DR, et al. Actions of the D1
agonists A-77636 and A-89929 on locomotion and dyskinesia in
MPTP-treated L-dopa primed common marmosets. Psychophar-
macology (Berl) 1999;142:51– 60.
16. Tsui JK, Wolters EC, Peppard RF, Calne DB. A double-blind,
placebo-controlled, dose-ranging study to investigate the safety
Movement Disorders, Vol. 19, No. 7, 2004
752 G. LINAZASORO
and efficacy of CY 208243 in patients with Parkinson’s disease.
Neurology 1989;39:856 – 858.
17. Taylor JR, Lawrence MS, Redmond DE, et al. Dihydrexidine, a
full dopamine D1 agonist, reduces MPTP-induced parkinsonism
in monkeys. Eur J Pharmacol 1991;199:389 –391.
18. Blanchet PJ, Fang J, Gillespie M, et al. Effects of the full
dopamine D1 receptor agonist dihydrexidine in Parkinson’s dis-
ease. Clin Neuropharmacol 1998;21:339 –343.
19. Shiosaki K, Jenner P, Asin KE, et al. ABT-431: the diacetyl
prodrug of A-86929, a potent and selective dopamine D1 receptor
agonist: in vitro characterization and effects in animal models of
Parkinson’s disease. J Pharmacol Exp Ther 1996;276:150 –160.
20. Rascol O, Blin O, Thalamas C, et al. ABT-431, a D1 receptor
agonista prodrug, has efficacy in Parkinson’s disease. Ann Neurol
1999;45:736 –741.
21. Rascol O, Nutt J, Blin O, et al. Induction by D1 receptor agonista
ABT-431 of dyskinesia similar to levodopa in patients with
Parkinson’s disease. Arch Neurol 2001;58:249 –254.
22. Gulwadi AG, Korpinen CD, Mailman RB, et al. Dinapsoline:
characterization of a D1 dopamine receptor agonist in a rat model
of Parkinson’s disease. J Pharmacol Exp Ther 2001;296:338 –
344.
23. Brücke T, Bankeiwicz K, Harvey-White J, Kopin I. The partial
dopamine agonist terguride in the MPTP-induced hemiparkinso-
nian monkey model. Eur J Pharmacol 1988;148:445– 448.
24. Baronti F, Mouradian MM, Conant KE, et al. Partial dopamine
agonist therapy of levodopa-induced dyskinesias. Neurology
1992;42:1241–1243.
25. Nichols NF, Cimini MG, Haas JV, et al. PNU-96391A
(OSU6162) antagonizes the development of behavioural sensiti-
zation induced by dopamine agonists in a rat model for Parkin-
son’s disease. Neuropharmacology 2002;43:817– 824.
26. Bézard E, Ferry S, Mach U, et al. Attenuation of levodopa-
induced dyskinesia by normalizing dopamine D3 receptor func-
tion. Nat Med 2003;9:762–777.
27. Silverdale MA, Millan MJ, Crossman AR, Brotchie JM. Dopa-
mine D3 receptor blockade, and not stimulation, is associated
with anti-parkinsonian activity in the MPTP-lesioned, non-human
primate model of Parkinson’s disease. Soc Neurosci Abstr 2002:
S63.001.
28. Richardson PJ, Kase H, Jenner PG. Adenosine A2a receptor
antagonists as new agents for the treatment of Parkinson’s dis-
ease. Trends Pharmacol 1997;18:338 –344.
29. Hauser R, Hubble JP, Truong DD and the Istradephylline US-001
Study Group. Randomized trial of the Adenosine A2A receptor
antagonist istradephylline in advanced PD. Neurology 2003;61:
297–303.
30. Sherzai A, Bara-Jiménez W, Sherzai A, Dimitrova T, et al.
Adenosine A2a antagonist treatment of Parkinson’s disease. Neu-
rology 2003;61:293–296.
31. Mally J, Stone TW. Theophylline, a not selective adenosine
antagonist in Parkinson’s disease. J Pharm Pharmacol 1994;46:
515–517.
32. Kulisevsky J, Barbanoj M, Gironell A, et al. A double-blind
crossover, placebo-controlled study of the adenosine A2A antag-
onist theophylline in Parkinson’s disease. Clin Neuropharmacol
2002;25:25–31.
33. Henry B, Fox SH, Peggs D, et al. The alpha 2-adrenergic receptor
antagonist idazoxan reduces dyskinesia and enhances anti-parkin-
sonian actions of levodopa in the MPTP-lesioned primate model
of Parkinson’s disease. Mov Disord 1999;14:744 –753.
34. Hill MP, Savola JM, Merivouri H, et al. Potential applications of
the alpha2 adrenergic antagonist, JP-1730, in reducing the prob-
lem of L-dopa-induced motor fluctuations in Parkinson’s disease:
studies in the rodent and primate models of Parkinson’s disease.
Mov Disord 2002;17(Suppl. 5):S71.
35. Gómez Mancilla B, Bédard PJ. Effects of nondopaminergic drugs
on L-dopa-induced dyskinesias in MPTP monkeys. Clin Neurop-
harmacol 1993;16:418 – 427.
Movement Disorders, Vol. 19, No. 7, 2004
36. Manson AJ. Iakovidou E, Lees AJ. Idazoxan is ineffective for
levodopa-induced dyskinesias in Parkinson’s disease. Mov Dis-
ord 2000;15:336 –337.
37. Rascol O, Arnulf I, Peyro-Saint PH, et al. Idazoxan, an alpha-2
antagonist, and L-DOPA-induced dyskinesias in patients with
Parkinson’s disease. Mov Disord 2001;16:708 –713.
38. Fox SH, Henry B, Hill MP, et al. Neural mechanisms underlying
peak-dose dyskinesia induced by levodopa ans apomorphine are
distinct: evidence from the effects of the alpha-2 adrenoceptor
antagonist idazoxan. Mov Disord 2001;16:642– 650.
39. Pact V, Giduz T. Mirtazapine treats resting tremor, essential
tremor and levodopa-induced dyskinesias. Neurology 1999;53:
1154.
40. Calon F, Goulet M, Blanchet PJ, et al. Levodopa or D2 agonist
induced dyskinesia in MPTP monkeys: correlation with changes
in dopamine and GABAa receptors in the striatopallidal complex.
Brain Res 1995;680:43–52.
41. Calon F, Morissette M, Goulet M, et al. Chronic D1 and D2
dopaminomimetic treatment of MPTP-denervated monkeys: ef-
fects on basal ganglia GABA (A)/benzodiazepine receptor com-
plex and GABA content. Neurochem Int 1999;35:81–91.
42. Calon F, Di Paolo T. Levodopa response motor complications:
GABA receptors and preproenkephalin expression in human
brain. Parkinsonism Relat Disord 2002;8:449 – 454.
43. Calon F, Morisette M, Rajput AH, et al. Changes in GABA
receptors and dopamine turnover in the postmortem brains of
parkinsonians with levodopa-induced motor complications. Mov
Disord 2003;18:241–253.
44. Lozano AM, Carella F. Physiologic studies in the human brain in
movement disorders. Parkinsonism Relat Disord 2002;8:455–
458.
45. Lees AJ, Shaw KM, Stern GM. Baclofen in Parkinson’s disease.
J Neurol Neurosurg Psychiatry 1978;41:707–708.
46. Turjanski N, Lees AJ. Gamma vinyl GABA in the treatment of
levodopa-induced dyskinesias. J Neurol Neurosurg Psychiatry
1992;55:413.
47. Ziegler M, Fournier V, Bathien N, et al. Therapeutic response to
progabide in neuroleptic- and L-dopa-induced dyskinesias. Clin
Neuropharmacol 1987;10:238 –246.
48. Price PA, Parkes JD, Marsden CD. Sodium valproate in the
treatment of levodopa-induced dyskinesias. J Neurol Neurosurg
Psychiatry 1978;41:702–706.
49. Reeves AL, So EL, Sharbrough FW, Krahn LE. Movement dis-
orders associated with the use of gabapentin. Epilepsia 1996;37:
988 –990.
50. Van Blercom N, Lasa A, Verger K, et al. Effects of gabapentin on
the motor response to levodopa: a double-blind, placebo-con-
trolled, crossover study in patients with complicated Parkinson’s
disease. Clin Neuropharmacol 2004 (in press).
51. Olson WL, Gruenthal M, Mueller ME, Olson WH. Gabapentin
for parkinsonism: a double-blind, placebo-controlled, crossover
trial. Am J Med 1997;102:60 – 66.
52. Chaná P, de Marinis A, Barrientos N. Gabapentin and motor
fluctuations in Parkinson’s disease. Mov Disord 1997;12:608.
53. Ondo W, Hunter C. Flumazenil, a GABA antagonist may im-
prove features of Parkinson’s disease. Mov Disord 2003;18:683–
684.
54. Fox SH, Brotchie JM. 5-HT2C receptor binding is increased in
the substantia nigra pars reticulata in Parkinson’s disease. Mov
Disord 2000;15:1064 –1069.
55. Tecott LH, Sun LM, Akana SF, et al. Eating disorder and epilepsy
in mice lacking 5-HT2c serotonin receptor. Nature 1995;374:
542–546.
56. Fox SH, Moser B, Brotchie JM. Behavioural effects of 5-HT2C
receptor antagonism in the substantia nigra zona reticulata of the
6-hydroxydopamine-lesioned rat model of Parkinson’s disease.
Exp Neurol 1998;151:35– 49.
 FAILURES OF SYMPTOMATIC TREATMENTS FOR PD
57. Eberle-Wang K, Lucki I, Chesselet MF. A role for the subtha-
lamic nucleus in 5-HT2c-induced oral dyskinesia. Neuroscience
1996;72:117–128.
58. Meco G, Marini S, Lestingi L, et al. Controlled single-blind
cross-over study of ritanserin and placebo in L-dopa-induced
dyskinesias in Parkinson’s disease. Curr Ther Res 1988;43:262–
270.
59. Durif F, Vidailhet M, Bonnet AM, et al. Levodopa-induced
dyskinesias are improved by fluoxetine. Neurology 1995;45:
1855–1858.
60. Richard IH, Maughn A, Kurlan R. Do serotonin reuptake inhib-
itor antidepressants worsen Parkinson’s disease? A retrospective
case series. Mov Disord 1999;14:155–157.
61. Kulisewski J, Campos V, Martı́nez P, Sancho J. Treatment of
depression with sertraline does not worsen motor symptoms in
Parkinson’s disease. Neurology 2003;60:A291.
62. Bibbiani F, Oh JD, Chase TN. Serotonin 5-HT1A agonist im-
proves motor complications in rodent and primate parkinsonian
models. Neurology 2002;58:P05.092.
63. Bara W, Sherzai A, Bibbiani F, et al. 5-HT1a agonist effects in
advanced Parkinson’s disease. Neurology 2003;60:A507.
64. Linazasoro G. Mecanismos glutamatérgicos en la enfermedad de
Parkinson. Neurologia 2000(Suppl. 6):21–30.
65. Klockgether T, Turski L. Toward an understanding of the role of
glutamate in experimental parkinsonism: agonist-sensitive sites in
the basal ganglia. Ann Neurol 1993;34:585–593.
66. Kaur S, Starr MS. Differential effects of intrastriatal and intrani-
gral injections of glutamate antagonists on motor behaviour in the
reserpine-treated rat. Neuroscience 1997;76:345–354.
67. Ikonomidou C, Turski L. Why did NMDA receptor antagonists
fail clinical trials for stroke and traumatic brain injury? Lancet
Neurol 2002;1:383–386.
68. Chase TN, Oh JD. Striatal mechanisms and pathogenesis of
parkinsonian signs and motor complications. Ann Neurol 2000;
47(Suppl. 1):S122–S130.
69. Nash JE, Brotchie JM. Characterization of striatal NMDA recep-
tors involved in the generation of parkinsonian symptoms: intra-
striatal microinjection studies in the 6-OHDA-lesioned rat. Mov
Disord 2002;17:455– 466.
70. Hill MP, McGuire SG, Crossman AR, et al. The mGluR5 receptor
antagonist SIB-1830 reduces levodopa-induced dyskinesia in the
MPTP-lesioned primate model of Parkinson’s disease. Soc Neu-
rosci Abstr 2001;25:200.
71. Montastruc JL, Rascol O, Senard JM, et al. A pilot study of
N-methyl-D-aspartate (NMDA) antagonist in Parkinson’s dis-
ease. J Neurol Neurosurg Psychiatry 1992;55:630 – 631.
72. Greenamyre JT, Eller RV, Zhang Z, et al. Antiparkinsonian
effects of remacemide hydrochloride, a glutamate antagonist, in
rodent and primate models of Parkinson’s disease. Ann Neurol
1994;35:655– 661.
73. Parkinson Study Group. A randomized, controlled trial of rema-
cemide for motor fluctuations in Parkinson’s disease. Neurology
2001;56:455– 462.
74. Starr MS, Starr BS, Kaur S. Stimulation of basal and L-dopa-
induced motor activity by glutamate antagonists in animal models
of Parkinson’s disease. Neurosci Biobehav Rev 1997;21:437–
446.
75. Merims D, Ziv I, Djaldetti R, et al. Riluzole for levodopa-induced
dyskinesias in advanced Parkinson’s disease. Lancet 1999;353:
1764 –1765.
76. Vergahen Metman L, Blanchet PJ, van den Munckhof P, et al. A
trial of dextromethorphan in parkinsonian patients with motor
response complications. Mov Disord 1998;13:414 – 417.
77. Shinotoh H, Vingerhoets FJG, Lee CS, et al. Lamotrigine trial in
idiopathic parkinsonism: a double-blind, placebo-controlled,
crossover study. Neurology 1997;48:1282–1285.
78. Vergahen Metman L, Del Dotto P, van den Munckhof P, et al.
Amantadine as treatment for dyskinesias and motor fluctuations
in Parkinson’s disease. Neurology 1998;50:1323–1326.
753
79. Verhagen Metman L, Del Dotto P, LePoole K, et al. Amantadine
for levodopa-induced dyskinesias: a 1-year follow-up study. Arch
Neurol 1999;56:1383–1386.
80. Sañudo-Peña MC, Tsou K, Walker JM. Motor actions of canna-
binoids in the basal ganglia output nuclei. Life Sci 1999;65:703–
713.
81. Brotchie JM. CB1 cannabinoid receptor signalling in Parkinson’s
disease. Curr Opin Pharmacol 2003;3:54 – 61.
82. Sañudo-Peña MC, Walker JM. Effects of intrapallidal cannabi-
noids on rotational behaviour in rats: interactions with the dopa-
minergic system. Synapse 1998;28:27–32.
83. Sañudo-Peña MC, Walker JM. Role of the subthalamic nucleus
on cannabinoid actions in the substantia nigra of the rat. J Neu-
rophysiol 1997;77:1635–1638.
84. Sañudo-Peña MC, Patrick SL, Khen S, et al. Cannabinoids effect
in basal ganglia in a rat model of Parkinson’s disease. Neurosci
Lett 1998;248:171–174.
85. Ruiz Ortega JA, Linazasoro G, Bilbao G, et al. Effect of WIN
55,512 on subthalamic nucleus and substantia nigra zona com-
pacta neurons in anaesthetized rats (in preparation).
86. Ruiz Ortega JA, Linazasoro G, Ugedo L. Effects of systemic and
intrasubthalamic dopaminergic drugs administration on the neu-
rophysiological activity of the subthalamic nucleus of rats with
moderate and severe lesions of the nigrostriatal pathway. Eur
J Neurosci (submitted).
87. Sieradzan KA, Fox SH, Hill MP, et al. Cannabinoids reduce
levodopa-induced dyskinesias in Parkinson’s disease: a pilot
study. Neurology 2001;57:2108 –2111.
88. Jenner P. Factors influencing the onset and persistence of dyski-
nesia in MPTP-treated primates. Ann Neurol 2000;47(Suppl.
1):90 –104.
89. Henry B, Brotchie JM. Potential of opioid antagonists in the
treatment of levodopa-induced dyskinesias in Parkinson’s dis-
ease. Drugs Ageing 1996;9:149 –158.
90. Piccini P, Weeks RA, Brooks DJ. Alterations in opioid receptor
binding in Parkinson’s disease patients with levodopa-induced
dyskinesias. Ann Neurol 1997;42:720 –726.
91. Sandyk R, Snider SR. Naloxone treatment of L-dopa-induced
dyskinesias in Parkinson’s disease. Am J Psychiatry 1986;143:
118.
92. Manson AJ, Katzenschlager R, Hobart J, et al. High dose nal-
trexone for dyskinesias induced by levodopa. J Neurol Neurosurg
Psychiatry 2001;70:554 –556.
93. Hughes NR, McKnight AT, Woodruff GN, et al. Kappa-opioid
receptor agonists increase locomotor activity in the monoamine-
depleted rat model of parkinsonism. Mov Disord 1998;13:228 –
233.
94. Krack P, Hamel W, Mehdorm HM, Deuschl G. Surgical treatment
of Parkinson’s disease. Curr Opin Neurol 1999;12:417– 425.
95. Vitek JL. Physiology of hypokinetic and hyperkinetic movement
disorders: model for dyskinesia. Ann Neurol 2000;47(Suppl. 1):
S131–S140.
96. Van Horn G, Schiess MC, Soukup VM. Subthalamic deep brain
stimulation: neurobehavioural concerns. Arch Neurol 2001;58:
1205–1206.
97. Grupo de estudio de estimulación cerebral profunda en la enfer-
medad de Parkinson. Estimulación palidal y subtalámica en la
enfermedad de Parkinson: Enseñanzas de los malos resultados.
Neurologia 2001;16:298 –302.
98. Linazasoro G. Subthalamic deep brain stimulation for advanced
Parkinson’s disease: all that glitters is not gold. J Neurol Neuro-
surg Psychiatry 2003;74:827.
99. Dunnett SB, Björklund A. Prospects for new restorative and
neuroprotective treatments in Parkinson’s disease. Nature 1999;
399(Suppl.):A32–A39.
100. Freed CR, Greene PE, Breeze RE, et al. Transplantation of
embryonic dopamine neurons for severe Parkinson’s disease.
N Eng J Med 2001;344:710 –719.
Movement Disorders, Vol. 19, No. 7, 2004
754 G. LINAZASORO
101. Olanow CW, Goetz CG, Kordower JH, et al. A double-blind
controlled trial of bilateral fetal nigral transplantation in Parkin-
son’s disease. Ann Neurol 2003;54:403– 414.
102. Watts RL, Raiser CD, Stover NP, et al. Stereotaxic intrastriatal
implantation of retinal pigment epithelial cells attached to micro-
carriers in advanced Parkinson’s disease patients: a pilot study in
six patients. Parkinsonism Relat Disord 2001;7:S87.
103. Gill SS, Patel NK, Hotton GR, et al. Direct brain infusion of
glial-derived neurotrophic factor in Parkinson disease. Nat Med
2003;9:589 –595.
104. Bezard E, Imbert C, Gross CE. Experimental models of Parkin-
son’s disease: from the static to the dynamic. Rev Neurosci
1998;9:71–90.
105. Bezard E. Neuroprotection for Parkinson’s disease: a call for
clinically driven experimental deign. Lancet Neurol 2003;2:393.
106. Pifl C, Schingnitz G, Hornykiewicz O. The neurotoxin MPTP
does not reproduce in the rhesus monkey the interregional pattern
of striatal dopamine loss typical of human idiopathic Parkinson’s
disease. Neurosci Lett 1988;92:228 –233.
107. Kish SH, Shannak K, Hornykiewicz O. Uneven pattern of dopa-
mine loss in the striatum of patients with idiopathic Parkinson’s
disease. N Eng J Med 1988;318:876 – 880.
108. Jenner P, Al-Barghoulthy G, Smith L, et al. Initiation of entaca-
pone with levodopa further improves antiparkinsonian activity
Movement Disorders, Vol. 19, No. 7, 2004
and avoids dyskinesia in the MPTP model of Parkinson’s disease.
Neurology 2002;58:P05.096.
109. Dickson DW. Neuropathology of parkinsonian disorders. In: To-
losa E, Jankovic J, editors. Parkinson’s disease and movement
disorders. 4th ed. New York: Lippincott Williams and Wilkins;
2002. p 256 –270.
110. Rye D, DeLong MR. Time to focus on the locus. Arch Neurol
2003;60:320.
111. Augood SJ, Hollingsworth ZR, Standaert DG, et al. Localization
of dopamine markers in the human subthalamic nucleus. J Comp
Neurol 2000;421:247–255.
112. Aizman O, Brismar H, Uhlén P, et al. Anatomical and physio-
logical evidence for D1 and D2 dopamine receptor colocalization
in neostriatal neurons. Nat Neurosci 2000;3:226 –230.
113. Beal MF. Experimental models of Parkinson’s disease. Nat Rev
Neurosci 2001;2:325–332.
114. Dawson TM. New animal models for Parkinson’s disease. Cell
2000;101:115–118.
115. Gorba T, Allsopp TE. Pharmacological potential of embryonic
stem cells. Pharmacol Res 2003;47:269 –278.
116. Sturla LM, Fernández-Teijeiro A, Pomeroy SL. Application of
microarrays to neurological disease. Arch Neurol 2003;60:676 –
683.