sign in sign up
<<
Home , Adenosine A2A receptor, Istradefylline, Methysergide, Pramipexole, Rotigotine, Sarizotan

REVIEWS

Novel pharmacological targets for the treatment of Parkinson’s disease Anthony H. V. Schapira*,‡, Erwan Bezard§, Jonathan Brotchie||, Frédéric Calon¶, Graham L. Collingridge#, Borris Ferger**, Bastian Hengerer**, Etienne Hirsch‡‡,§§, Peter Jenner||||, Nicolas Le Novère¶¶, José A. Obeso##, Michael A. Schwarzschild***, Umberto Spampinato‡‡‡ and Giora Davidai§§§ Abstract | deficiency, caused by the degeneration of nigrostriatal , is the cause of the major clinical motor symptoms of Parkinson’s disease. These symptoms can be treated successfully with a range of drugs that include levodopa, inhibitors of the enzymatic breakdown of levodopa and dopamine delivered by oral, subcutaneous, transcutaneous, intravenous or intra-duodenal routes. However, Parkinson’s disease involves degeneration of non-dopaminergic neurons and the treatment of the resulting predominantly non-motor features remains a challenge. This review describes the important recent advances that underlie the development of novel dopaminergic and non- dopaminergic drugs for Parkinson’s disease, and also for the motor complications that arise from the use of existing therapies.

Bradykinesia Parkinson’s disease (PD) is a multicentric neurodegen- those aimed at managing or preventing the onset of motor Abnormally slow voluntary erative disease with a prevalence of approximately 1 in complications), the identification of non-dopaminergic movements. 300. Clinical features at presentation include the asym- drugs for symptomatic improvement and the discovery metric onset of bradykinesia, rigidity and tremor. These of compounds to modify the course of PD. Lewy body Abnormal aggregates of are the result of the loss of dopaminergic neurons in the protein (predominately pars compacta, which causes a con- Dopaminergic drugs α-synuclein) that develop sequent reduction of dopamine levels in the Dopamine-replacement therapy has dominated the treat- inside cells and displace (FIG. 1). However, additional neuronal fields and neuro- ment of motor symptoms of PD since the early 1960s. The other cell components. transmitter systems are also involved in PD, including effects are predictable (as are the side effects) and none of the locus coeruleus, dorsal motor nucleus, substantia the more recently introduced synthetic dopamine agonists Involuntary writhing innominata, the autonomic nervous system and the has surpassed the clinical benefit derived from levodopa movements affecting head, cerebral cortex (FIG.2). Consequently non-, (!-DOPA)2,3 (TABLE 1). Most recently, non-oral delivery has trunk and limbs. serotinergic and cholinergic neurons are also lost. provided more long-lasting anti-Parkinsonian activity This loss results in symptoms that include cognitive through the subcutaneous or intravenous infusion of decline, sleep abnormalities and depression, as well as and technology with gastrointestinal and genitourinary disturbances. These or lisuride4,5. However, despite the many

*University Department of ‘non-motor’ features progress and come to dominate dopaminergic agents currently available, the search for Clinical Neurosciences, Royal the later stages of PD. Although the clinical conse- novel approaches based on dopamine-replacement therapy Free and University College quences of non-dopaminergic neuronal involvement continues. The multiplicity of dopamine receptors in Medical School, University usually become apparent some years after diagnosis, the brain offers a range of potential targets, but so far College London, Rowland Hill Street, London NW3 2PF, UK. there is debate about the sequence in which PD pathol- exploitation of drugs acting on specific subtypes ‡Institute of Neurology, ogy develops. The distribution of Lewy body formation has been disappointing. Most currently used drugs only 6 Queen Square, University might include non-dopaminergic areas at an early activate D2 and D3 dopamine receptors and no major College London, London 1 stage ; however, it remains to be shown that cell death advance has been made in producing D1 dopamine ago- WC1N 3BG, UK. occurs in these areas before the substantia nigra. nists, a known target for anti-Parkinsonian agents. Partial Correspondence to A.H.V.S.: e-mail:a.schapira@medsch. The three main strategic developments that have led to D2 dopamine agonists are being developed as they might ucl.ac.uk progress in the medical management of PD have focussed treat the motor symptoms of Parkinson’s disease while doi:10.1038/nrd2087 on improvements in dopaminergic therapies, (including suppressing both and dyskinesia7.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 5 | OCTOBER 2006 | 845 REVIEWS

Author addresses were among the first drugs used in PD, and were intended to correct the imbalance § CNRS UMR 5543, Universite Victor Segalen–Bordeaux 2, 146, rue Leo Saignat, 33076 between dopamine and levels. Although Bordeaux cedex, France. these drugs do produce some beneficial effects on PD ||Toronto Western Research Institute, Toronto Western Hospital, 399 Bathurst, MC symptoms, they are associated with adverse cogni- 11-419, Toronto, Ontario M5T 2S8, Canada. 17 ¶Molecular Endocrinology and Oncology Research Centre, Centre Hospitalier de tive effects . Many cholinesterase inhibitors, such l’Université Laval Research Centre (CHUL), Quebec (QC), Canada, G1V 4G2 and Faculty as rivastigmine (Exelon; ) and donepezil of Pharmacy, Laval University, Quebec G1K 7P4, Canada. (Aricept; Eisai/Pfizer), have been tested to counter- #MRC Centre for Synaptic Plasticity, University of Bristol, University Walk, Bristol, act PD dementia, and have been found to improve BS8 1TD, UK. cognition18–21. However, they sometimes display **Boehringer Ingelheim Pharma GmbH & Co. KG, 88397 Biberach, Germany. mixed effects on motor function22. It might be that ‡‡INSERM, UMR 679, Neurology and Experimental Therapeutics, Paris, France. co-treatment using a combination of anticholinergics §§ University Pierre et Marie Curie, Hôpital de la Pitié-Salpétrière, Paris, France. and anticholinesterases would correct acetylcholine |||| Neurodegenerative Disease Research Centre, School of Health and Biomedical deficits while counteracting the hypersensitivity of Sciences, King’ s College, London SE1 1UL, UK. cortical muscarinic receptors. ¶¶Computational Neurobiology, EMBL-EBI. Wellcome-Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK. Nicotinic receptors are not only highly expressed 23 ##Department of Neurology and Neuroscience Centre, Clinica Universitaria and Medical on dopaminergic neurons , but also in the cortex and School, Universidad de Navarra and CIMA, Avenida de Pio XII, Pamplona 31008, Spain. thalamus. has been found to protect against ***Department of Neurology, Massachusetts General Hospital, MassGeneral Institute degeneration in various PD models24,25. In addition, an for Neurodegenerative Disease, 114 16th Street, Boston, Massachusetts 02129, USA. inverse association between smoking and PD has been ‡‡‡14 Unité Mixte de Recherche-Centre National de la Recherche Scientifique, consistently demonstrated26. However, nicotine itself (UMR-CNRS) 5541, Université Victor Segalen Bordeaux 2, 146 rue Léo Saignât, B.P. 31, has no anti-Parkinsonian effect and the mechanism by 33076 Bordeaux Cedex, France. which smoking might confer protection against PD is §§§ Department of Cardiovascular Medicine and Neurology, Clinical and Scientific Affairs, not known. Boehringer Ingelheim, 900 Ridgeway Road, Ridgefield, Connecticut CT06877, USA. Serotoninergic drugs 5-HT receptors are crucial to motor control in health 27–33 Although !-DOPA is the most effective drug for the and disease . In PD, 5-HT1A, 5-HT1B, 5-HT2A and

control of motor symptoms, it also causes a high level 5-HT2C deserve special attention, particularly with of motor complications, particularly dyskinesias8. By respect to involvement in !-DOPA -induced dyskinesia contrast, dopamine agonists produce a lower incidence (LID). of involuntary movements, which seems to reflect their In monkeys given 1-methyl-4-phenyl-1,2,3,6-tet- longer duration of action and supports the concept of rahydropyridine (MPTP), which induces Parkinsonian 9,10 continuous dopaminergic stimulation . This has been symptoms, the 5HT1A receptor agonists exploited by the introduction of a range of long-acting (Merck)34 and 8-hydroxy-2-di-n-propylamino-tetralin dopamine agonists, by the use of !-DOPA in combina- (8-OH-DPAT)35 dramatically reduce LID. In clinical 11 tion with catechol-O-methyl transferase inhibitors and trials, sarizotan and another 5HT1A , , more recently by development of once-daily controlled- reduced LID36–38 and extended the duration of !-DOPA released formulations of drugs and the action37. However, at high doses, sarizotan can exacer- introduction of transdermal dopamine agonist delivery7. bate Parkinsonism36. This might reflect an interaction

Whether continuous dopaminergic stimulation explains with D2 dopamine receptors. Therefore in developing

all of the differences between !-DOPA and dopamine the next generation of 5-HT1A agonists, it might be

agonists is a matter of contention. There might instead be beneficial to remove D2 activity. The potential value of fundamental differences between the actions of !-DOPA developing such compounds is highlighted by recent

and those of dopamine agonists. Unlike the dopamine observations that 5-HT1A agonists might also be

agonist drugs, !-DOPA acts on both D1 and D2 receptor neuroprotective (E.B., unpublished observations).

families, affects multiple pharmacological targets (includ- Rodent studies suggest that antagonism of 5-HT2A 39 ing nonadrenaline and 5-hydroxytryptamine (5-HT; and 5-HT2C receptors can reduce LID either directly or ) receptors) and might act as a neuromodulator indirectly by allowing reduction in !-DOPA dosage40,41. 12 in its own right . Alternatively, dopamine agonists also In MPTP primates, blockade of all 5-HT2 receptor sub- have additional properties other than actions at dopamine types with reduces dyskinesia, though receptors, which could explain their lower risk for the with on Parkinsonism42. The atypical neu- development of motor complications. roleptics (Seroquel; AstraZeneca) and cloza-

pine are 5-HT2A and 5-HT2C antagonists, in addition to Cholinergic drugs being dopamine antagonists43,44. In MPTP primates, both Both the cortico–striato–thalamic loop and the nigro– quetiapine and can reduce LID without exacer- striatal system are largely innervated by cholinergic bating Parkinsonism45, though at higher doses clozapine afferents coming from the tegmentum, the septum and does worsen . In a clinical study, clozapine by cholinergic interneurons. Most cholinergic systems reduced LID without affecting Parkinsonian disability46. are affected in PD, such as muscarinic receptors13,14, However, in another clinical study, quetiapine failed to nicotinic receptors13,15 and choline transporters16. demonstrate any benefit on dyskinesia, though the doses

846 | OCTOBER 2006 | VOLUME 5 www.nature.com/reviews/drugdisc REVIEWS

48 Baseline 22 mo both 5-HT1A or 5-HT1B receptors . Although itself unten- able as a therapeutic agent, the MDMA molecule might represent a starting point for novel medicinal chemistry to develop new drugs that combine a propitious mix of 5-HT actions of benefit to PD.

Glutamate and GABA drugs Since the vast majority of pathways in the utilize glutamate and GABA (γ-amino butyric acid) as their respective excitatory and inhibitory neurotransmit- ters, these systems are obvious drug candidates. Indeed, 34 mo 46 mo there is already some evidence that N-methyl-"-aspar- tate (NMDA) receptor antagonists, such as , and , might reduce motor complications associated with !-DOPA therapy. Targeting these amino-acid receptor systems, although potentially very attractive, is fraught with complications. Firstly, there is the problem of regional selectivity given the ubiquitous nature of these receptor systems in the brain. The second problem is the potential for affecting normal basal ganglia function. For example, antagonizing NMDA receptors could offer neuroprotec- Figure 1 | A single-photon emission computerized tomographic (SPECT) scan tion and limit pathological plasticity but will interfere of density in a Parkinson’s disease patient shortly after with the normal function of these receptors as mediators diagnosis and serially for 46 months. There is asymmetric loss particularly in the posterior putamen, which progresses bilaterally over time. Levels of SPECT activity of high-frequency synaptic transmission and synaptic are colour-coded from low (black) to high (yellow/white). β-CIT, 2β-carboxymethoxy- plasticity (for example, long-term potentiation, long- 3β(4-iodophenyl). Adapted, with permission, from REF. 95  (2002) American term depression and depotentiation). Medical Association. A detailed understanding of the sites and mecha- nisms of action of the compounds can, however, be used to optimize and minimize potential side used were lower than those found to be effective in effects. (Axura; Merz) is an example of 47 monkeys . As with 5-HT1A agonists, it is likely that how an NMDA with specific prop- 51 the therapeutic window of currently available 5-HT2A/C erties can provide therapeutic potential . Because it is agents will be limited by antidopaminergic actions, a low-affinity and highly voltage-dependent NMDA though their use could pave the way for more selective receptor antagonist it can limit spurious NMDA recep- compounds in development (for example, ACP-103 tor activation without preventing the intense transient (Acadia)). activation that results from the coordinated synaptic

In MPTP primates, 5-HT1B receptor stimulation release of glutamate. In addition to the mechanism of 35,48 reduced LID . 5-HT1B agonists and partial agonists block, subtype-selective NMDA receptor antagonists are widely available for other indications and are gen- provide another potential therapeutic angle. In this erally unencumbered by direct dopaminergic effects. context, selective antagonists of NDMA receptor 2B 52,53 A 5-HT1B agonist/ capable of crossing (NR2B) might offer anti-Parkinsonian effects . the blood–brain barrier and with Interestingly, the high expression of NR2D receptors on similar to !-DOPA might be an excellent candidate for substantia nigra pars compacta neurons, but relatively a novel antidyskinetic agent. low expression of this receptor on many other neuro- The diversity of 5-HT receptors involved in PD raises nal types, makes these receptors an interesting target54. the issue of whether the most effective 5-HT-modulat- Lead compounds for developing NR2D-selective ing drugs for PD might be molecules that target a range antagonists are now available55. of receptors rather than highly selective ones. However, Targeting metabotropic glutamate (mGlu) receptors a non-selective increase in 5-HT transmission, such (mGluRs) is another interesting possibility. The eight as achieved with a selective serotonin-reuptake inhibi- receptor subtypes (mGlu1–8) have a varied distribution in tor (SSRI), does not reduce LID in either monkeys48 or the brain. Recent evidence suggests that profound altera- humans49. However, it would be desirable to discover a tions in depotentiation, a process that probably utilizes molecule combining two or more of the specific agonist mGlu receptors, could underlie dyskinesia56. Interesting and antagonist properties discussed above. The feasibility candidates for development are allosteric potentiators of of this approach is highlighted by findings that mirtazap- group III mGluRs. For example, it has been shown that

ine (Remeron; Organon), a molecule combining 5-HT1A N-phenyl-7-(hydroxylimino)cyclopropa[b]chromen-1a-

agonist and 5HT2A antagonist properties, can reduce LID carboxamide selectively activates mGlu4 receptors and Depotentiation 50 57 The reversal of long-term in patients , while in MPTP monkeys the non-selective markedly reverses -induced akinesia . Another potentiation back to agent 3,4 methylenedioxymethamphetamine prime target is the AMPA (α-amino-5-hydroxy-3-methyl- baseline levels. (MDMA, or ‘ecstasy’) can reduce dyskinesia by stimulating 4-isoxazole propionic acid) receptor, which mediates most

NATURE REVIEWS | DRUG DISCOVERY VOLUME 5 | OCTOBER 2006 | 847 848 R L E |

O V C T O I B E E R W

2 0 0 A P 6 S

|

V O L U M M p n C p D d ( s a i a l e A C in t r f g d F e n t a h e e n r r i u r e a o E i u a M M g o f d s e e s e c t v e l t h r c p

r u t o e l n u

i e e 5 y

m q m m

i N l o c a a n , P e d a e

r p

v r

u b a c c n f m e e i x A , o a a s i u e t

s i ; linical t a r t d

c l c e o r r r o

n s 2 s a e h i y

S u r k k e o i s n i . ; r s

i

s - t e o e

I | a e d e e m n G r T

e e

, a

c c S

l a n l s f d d s a b

h a

t ) o y o f e P a a n c , e u , o t e

r t , e

p ( i m n n r i l f

n h e a e t b n t r s ,

c e r

m t

a o

u f ; i e t y e i r o tr t s

t o DMN

p i a n n b

g S

h r r g e t v

a m o m o PPN r VT ; s a s q t e a l ials o N

e e d n u

RN RN a SN r A y e t A8 LC t f d t n u a

r n

o u e r

c R d , i A m n f r i m r

a t i e t t a n s n m k r i n E o e

i i a i l n s u

i a r

v c a g F

n o as c n s o l P p s t

d b

e l s

e l e t i , s u n t a a 1 e r r t

n e u i s s e r a

,

i e 1 r c o

l f = s i

p e t t a c n c i m r k r t s n l h p 9 n a n h

t

g o , e t o o i u d i

t e h a n

n e r i e A n y , u  m t r o m t

o c e g w d t

g r s

i v a e

( s l h n b =

i p s o i o p e e e ( o i d n n

a e i o 2 n e s i

a t p n f n i s a l

a s s a t t

h n t n 0 f v n a ( m

’ o h

t ial m n o

l

s f s h e s k t i

0 t a

t r a

g r o

i t f h h a d i ) a m o a 4 E ; c

e n i

v r f g ; e a o t s v i A t

m )

neur

a r - e s e a g i e S

s r n i

e w E i e v 2 e C ; o f e ; i o l n l C

d

f w t a l o P a o b a 0 p s y n s C e r t o

, A g m n

s b s ; e 0

e ) i

t c

( s t

p n . , e o e u P b h e

L u

v

o 7 o t p O n i p

k , f s r n s

i , t e l n s

, p = i

e u h a o p

5 o

a p n

n = t

n

8 G p

b

r c u r c a t r n h p

t . a a

e b

o o

o t a n k h a l S v s T e l Y o e l e h r o l l

t n v

e e s o c e i c i l v l

s u e h e n a K d i t n d i , i

r i o t e d b

d n t b c e s

l e a d h u e e

r

o ; I r

t e u t i

v r n

d i a

( i r d m u o o n - a E a g m b c a i e s i c c r 4 e v t ;

r o n C g ; h i a n l e p ;

e c h

: o n 7 u S w e

H

r t ts

t P c n ,

u . o m o e ;

d 2 e T i , . e o

I O e n

a m t o L p r , m N ) A 6 e i n

e in g n

l g

h d f l r m i G = 1 g n b o , t n

n i M o t

t

i b s t o s

e p e e t , s h l i e j t a a e P

u r a i t r e r p s n n p e P e ; a a j i n v h t b e

D

100 100 100 c v o a ( d b s

e Controls (%) Controls (%) Controls (%) o t

A i e e % r i d d t 50 50 50 n t e e ; r r n l r

c i . r

r h i f 0 0 0 o u A

g

, a a o

e a o a

o a p l e h l t P n

l f m r p ; l f

o , i a

t

a c

=

v M t n a p PU a a a t p t e p

c p n o m c h

e o h r o t s s b x a t o o s u e

r e e h =

e o s r i a r t

t c c l t r s c a s

a , e m h o r e e

s Cau r e t e h i ; m

s

n o

y a

m

n w t l

x i n p H b i a a o c d s f

t a k e ; u b g o t r o a o l

n

Y g a b e t

F

i d a c m y l w c s y , o n a r g n i SN PU

FC e C l

l l r t i o c h o s i

e l a i r e e r u

h i

e n

a l m , f y n e o d i u l

f y y d

s s e f s ) l n c ) p o t m P r f .

s ; Amg

. e Cau

HI

, Thalamus Cau u SI l m

STN GPi GP G i

r

l EC

PU o p A O o v

; e A P r t a a n l o

h l a h V y n L m r t l s o i P r m c l

l n a s ; h i e t o i e i u s

t POG T

l N e o g t p C ) h

e CIC a P ) a v

m t A , y i 5

a t n

h AL h i v , h x l e t l n e . e

y n

f a ,

p h c s e T

a l e

p i f r m l r v o c

m y

s , c e o e v e h i r r

e VT r e SN l n e c i HI e

- h a c d r e

i e e e n u s c a n e

a n t o , r t o e f u

e A s n s r e e t g g i i e r t c t u l e a n o s ; l n ACC r Amg e

p x i t r

d

r o

i e ACC a e L n r a c d u s w

c a b a ; d t c l

f l

o C d

s e l m y G s u p

l o r

e u t w t h t

f

a u

n , l p e h i l h r p a r P d o

o ACC

l i e v l w f e u s g HY g e n p r s e n o e a f i n p HI t e o

5 n e c . m

m r d

t , s c a e

o c 9 n o

t j h v g e c h e t . a u

r l e a o a m a ; e n

h o w t l x p

o Amg

n o s r c t F C FC m r t i e n HY t t u

e h w g i o d t u t m t a c i e d i d t I n o i

a i s e e n r C n y o a a o n

r b

d

t e e g b t r

e CIC s n s a l e l o n a h e , . a

SN l , FC . ,

o y y r a l

n e a c

r c

l l

e G w

s O r n f

i s u s r n t l l i o

u p e a

s a c

y n e e h t e e s h l A Amg m i c t a f c g e e ,

l a c g e EC x s LC u a

e i

k l t t u i c t m . t / t e B g u c t g g r h o

o

h i s r A a e h s o u o o m k l a e o A n e r

; a i e n

d i t a s

y n a SC SC n n LC a s v n l P s n r r e

t ; o

r a ; r .

f i a i

b g s U

r a a c

R n e e e o g i g A n P p n i e u l v o n t w t

,

N f r d i a t i a

8 a e s n Cer c

l c

p e RN AP d r e a e o d b r m s l , e , t

t e a

o

s

u y

d k f / r r o e A e

s o r o ( p l a

a d i w t g e i i u ,

x p i t n 8 l

n a n n

p t w r e n r HY t c l ; b a

o o m u

e s t e h d s a t e g i m u h o b g r h i r t c v e p

e o

s h s e n n d u e i e i

a t n

t n

SC n s

i h

’ s i p i o i s ; s

s t e

l n

a

e c r s

- -

REVIEWS

Table 1 | Current drugs for Parkinson’s disease Drug Advantages Disadvantages Levodopa (!-dopa) + dopa • Probably the most potent dopaminergic drug • Motor complications decarboxylase inhibitor for symptom relief (cumulative risk 10% per • Generally well tolerated annum) Catechol-O-methyl transferase • Increase levodopa half-life • can cause inhibitors, for example, • Reduce ‘off’ time damage. , tolcapone • Diarrhoea dopamine agonists • Good efficacy • Increased risk of , (for example, , • Delay onset of motor complications , , , • Generally well tolerated peripheral oedema and Non-ergot dopamine agonists • Once-a-day preparations available with some behavioural changes for example, , • Transdermal patch for rotigitine • Cardiac valve fibrosis with , rotigitine • Theoretical neuroprotective action ergot drugs • Some action with pramipexole • Improve motor features in early and late • Relatively mild efficacy inhibitor; ; disease • Selegiline metabolized to • Easy to use, once-a-day — potential • Well tolerated cognitive effects • Theoretical neuroprotective effect Amantadine • Mild anti-Parkinsonian effect • Cognitive disturbances • Improves • Peripheral oedema • Livedo reticularis Anticholinergics • Mild anti-Parkinsonian effect • Limited by side effects such as confusion

60,61 LID , targeting GABA systems could provide a com- Third, the A2AR might have an important role in plementary strategy, although again subtype-selective the underlying neurodegenerative process, suggesting

compounds would seem to be the way forward. that A2A antagonists possess neuroprotective proper- ties. Epidemiological data linking the consumption

Adenosine A2A receptor antagonists of (a non-specific antagonist) with

Several distinctive features of the a reduced risk of developing PD have converged with

(A2AR) have made its antagonism a leading candidate laboratory studies showing that caffeine and more 62,63 strategy for the improved treatment of PD . First, specific A2A antagonists protect against dopaminergic and perhaps uniquely among currently pursued non- toxicity in vivo69,70. Another potential disease-

dopaminergic targets in PD research, A2ARs in the cen- modifying benefit of A2A antagonism — preventing the tral nervous system are relatively selectively expressed in development of LID in PD — has been proposed based the striatum, which is innervated by the dopaminergic on rodent and monkey studies71,72. These findings have

nigrostriatal neurons lost in PD. Moreover, within the further encouraged clinical trials that will investigate A2A 62 striatum A2ARs are largely restricted to the subset of antagonists in early PD .

medium spiny output neurons that project to the globus Lastly, the recent discovery that the A2AR can form 64 pallidus and co-express dopamine D2 receptors . This functional heteromeric receptor complexes with other

discreet anatomical localization reduces the liability of G-protein-coupled receptors such as D2R and the

A2A antagonists for adverse CNS effects such as those mGlu5 receptor has suggested new opportunities for that limit the usefulness of current non-dopaminergic leveraging the multiple potential anti-Parkinsonian 73 ( and antiglutamatergic) agents in PD. benefits of A2A antagonists . For example, combin-

Second, A2A antagonists consistently reverse Parkin- ing A2A and mGlu5 blockade synergistically reverses sonian motor deficits in all preclinical models of PD, Parkinsonian deficits in rodents74,75. On the other hand, and do so without inducing or exacerbating dyskinesias enthusiasm for these agents in PD is tempered by the in non- primate models63,65. This symptomatic possibility that they might produce untoward CNS or

effect can be explained by blockade of the A2AR on systemic effects. On balance, A2A antagonism represents

the D2R-coexpressing striatopallidal neurons, which one of the most realistic and promising therapeutic inhibits their release of GABA in the globus pallidus, candidates for PD. ultimately leading to enhanced motor function through the so-called indirect motor pathway of the basal ganglia66. drugs

Initial human studies with an A2A antagonist have indeed Recognition of the enhanced opioid peptide trans- demonstrated modest symptomatic improvements, mission in the striatum of animal models and PD Indirect motor pathway apparently without increasing troublesome dyskinesias patients with LID motor complications has raised the Globus pallidus-mediated in moderately advanced patients already experiencing possibility of controlling these by targeting opioid control of motor activity. motor fluctuations62,63,67,68 (TABLE 2). transmission in the basal ganglia76. Interestingly,

NATURE REVIEWS | DRUG DISCOVERY VOLUME 5 | OCTOBER 2006 | 849 REVIEWS

Table 2a | Selected drugs in Phase II–IV study for PD or novel indications in PD studies in the brain of monkeys and human patients have shown that non-dyskinesiogenic dopamino-mimetic Structure Drug Status treatments are associated with a correction of increased Cholinergic drugs preproenkephalin-A expression back to control levels10,77. O Donepezil Launched for So far, however, the use of non-selective opioid recep- O Alzheimer’s disease tor antagonists such as naloxone and naltrexone has provided contradictory or species-dependent results 78,79 O in non-human primates and inconclusive results in PD patients80,81. N The use of selective µ- and δ-receptor antagonists, but not of κ-receptor, however, looks more promising78. Almost none of the previous studies have considered Launched for opioid transmission at the peak of dyskinesia, but have N Alzheimer’s disease instead measured precursor levels a few hours or days after the actual presence of involuntary movements. The correlation between opioid expression and dyskinesias reported might therefore not be associated with dyskinesia OH O O per se, but rather with the supersensitive state of striatal dopamine receptors. If this hypothesis is correct, detection of preproenkephalin-A mRNA overexpression through Rivastigmine Launched for imaging technology might help predict the susceptibil- N Alzheimer’s disease O ity of patients to dyskinesia. In addition, although opioid O precursor levels have been extensively characterized, both the actual identity of their end products, the opioid pep- N tides, and the levels of opioid receptors are poorly known. Indeed, opioid precursors are potentially processed into a number of opioid peptides82, which then bind with Serotoninergic drugs various affinities to the three classes of opioid peptide receptors83,84. Considering the discrepancy between a N Sarizotan wide range of binding peptides and only three different H N receptors, and the complex anatomical distribution of O these receptors in the basal ganglia85, calls for developing F innovative approaches, possibly based on local delivery of small interfering RNA specific for the opioid precursors instead of classic pharmacological approaches. ACP-103 Phase II N Disease modification F O The development of drugs to slow or prevent the pro- H N N gression of PD might logically be thought to evolve from an improved understanding of the aetiology and O pathogenesis of PD. There have certainly been major advances in these areas over the past few years and the OH Quetiapine Launched for prospects for the introduction of ‘neuroprotective’ ther- NH+ O psychoses apies is much improved. Mutations in six different have been demonstrated in familial PD: α-synuclein, N parkin, UCHL1 (ubiquitin carboxyl-terminal N esterase L1), DJ1, PINK1 (PTEN induced putative kinase 1) and LRRK2 (leucine-rich repeat kinase 2). S Other mutations such as in NURR1 (Nur-related factor 1) and HTRA2 might also be associated with PD. Opioid drugs Although these are still considered uncommon causes Cyprodine Discovery/research of PD, understanding the biochemical consequences tool of their expression will provide important insights into 86 N the pathogenesis of the idiopathic disease . This understanding has led to the trial of anti-oxidants O CH 3 and pro-mitochondrial drugs in PD, with the latter showing some promise88. Activated microglial cells are present in the substantia nigra and it has been O O reported that these cells express tumour-necrosis H3C factor-α, interferon-γ, interleukin 1-β, the low-affinity immunoglobulin E receptor CD23, inducible nitric

850 | OCTOBER 2006 | VOLUME 5 www.nature.com/reviews/drugdisc REVIEWS

oxide, cyclooxygenase 2, complement 3 receptor and toxic compounds (6-hydroxydopamine (6-OHDA), increased ferritin89. A microglial reaction has also been MPTP, rotenone and annonacine), indicating that the observed in animal models of PD induced by several glial reaction is the consequence of dopamine degenera- tion, whatever the cause. Epidemiological studies have suggested that the use of anti-inflammatory drugs might Table 2b | Selected drugs in Phase II–IV study for PD or novel indications in PD reduce the risk for PD90. The identification of these bio- Structure Drug Status chemical abnormalities has helped the development of Dopaminergic drugs disease-modifying therapies. No drug has yet been shown to be neuroprotective Pramipexole Launched in PD, although several have been tested in clinical N trials. It is accepted that such a strategy will only be H2N N • 2 HCl • H2O successful if degeneration is ameliorated in multiple S H H systems, preventing the progression of both motor and non-motor features. The drugs that Rasagiline Launched have received most attention in relation to neuropro- tection include the dopamine agonists and monoamine oxidase (MAO) type B inhibitors, although others, NH including co- Q10, growth factors, anti-apoptotic agents and glutamate inhibitors, have also been the SLV-308 Phase III subjects of clinical trials in PD. N

H Cl Dopamine agonists. Dopamine agonists have demon- N strated neuroprotective properties in a wide range of O 91 in vitro and in vivo studies . The D2/D3 agonist prami- O pexole (Mirapex; Boehringer Ingelheim) has been shown N 92 H to protect non-human primates against MPTP toxicity . The basis for this protection is not understood, but it does Glutamatergic and GABAergic drugs not seem to depend on the presence of dopamine recep- 93 Me Talampanel Phase II tors . Dopamine agonists have been studied in clinical O O trials to assess potential disease modification with neuro- 94,95 N imaging markers as endpoints (FIG. 3). Both pramipex- O N Me ole and ropinirole showed slowed progression of their respective imaging endpoints compared with !-DOPA. These results could be interpreted as demonstrating neuroprotection. However, this interpretation is poten- tially confounded by a theoretical effect of the drugs on H2N imaging endpoints and could not discriminate between O NH2 E-2007 Phase III agonist protection and !-DOPA toxicity given the OH 96 HO absence of a placebo group . Further trials are underway N to address these issues. O MAO-B inhibitors N GYKI-47261 Discovery The MAO-B inhibitor selegiline (deprenyl) has dem- onstrated neuroprotective properties in a number of Cl N model systems relevant to PD97–103. The DATATOP N (Deprenyl And Tocopherol Antioxidant Therapy of Parkinson’s disease) trial demonstrated that selegiline could delay the introduction of !-DOPA in early PD 104 H2N by 9–12 months . The interpretation of this result was confounded by selegiline’s symptomatic effect. Rasagiline (Azilect; Teva) is a new potent MAO-B Adenosine A2A drugs inhibitor and has also shown neuroprotective effects OCH3 KW-6002 or Phase III O in in vitro and in vivo models of PD105–108. These effects OCH are independent of its MAO-B inhibition. The potential N N 3 (E) for rasagiline to delay progression in PD was evaluated O N N in the 12-month extension study TEMPO (TVP-1012 Early Monotherapy for Parkinson’s disease Outpatients; FIG. 4)109. At 12 months, those who started rasagiline 6 Structure unavailable SCH 420814 Phase II months before the delayed- start group maintained better clinical scores. The results cannot be explained by a Structure unavailable BIIB014/V2006 Phase I symptomatic effect alone and at face value represent

NATURE REVIEWS | DRUG DISCOVERY VOLUME 5 | OCTOBER 2006 | 851 REVIEWS

a 10 b 10 results111. The beneficial effects of GDNF could be a consequence of both protective and restorative proper- 0 0 ties. An early study of low-dose, monthly intraventricular –5 –5 injections of GDNF into PD patients did not produce any (39) clinical benefit, or any evidence of dopaminergic regen- –10 –10 (35) eration in the one patient that underwent autopsy112,113. odopa uptake

om baseline (33) A pilot study reported on direct putaminal infusion of –15 –15 114 fluor GDNF in five PD patients with advanced disease . In 18 *(39) –20 –20 contrast to intraventricular delivery of GDNF, putami- nal infusion was well tolerated and produced significant –25 –25 ‡(36) % change in fr clinical improvement and a reduction in dyskinesias.

% change in Ropinirole (68) Pramipexole Furthermore, fluorodopa PET scans showed a significant –30 –30 § !-DOPA (59) !-DOPA (32) increase in uptake in the putamen and substantia nigra at –35 –35 18 months. However, a larger controlled of 0 10 20 30 40 50 GDNF in PD was negative and terminated prematurely Scan interval (months) because of side effects115. This trial has been criticized for Figure 3 | Data from imaging studies in dopamine agonist neuroprotection trials its design and the dosage of GDNF used116. showing a slowing of progression in Parkinson’s patients in comparison to !-DOPA. a | Shows the percentage change in putamen F-DOPA emission p <0.001. Other compounds have been evaluated for disease- REF. 96 b | Shows the percentage change in putamen 2β-carboxymethoxy-3β modifying effect in PD with varying results (see (4-iodophenyl)tropane binding (*p = 0.005; ‡p = 0.001; §p = 0.03). Numbers in for a review). parentheses refer to patient numbers. Data adapted, with permission, from REF. 95 It has recently been suggested that early symptomatic  (2002) American Medical Association. treatment of PD might itself have some neuroprotective effect by modifying the compensatory mechanisms that maintain normal motor function in the pre-symptomatic an early disease-modifying effect. However, there are period during which nigral dopaminergic cell death is pro- Microglial reaction 117 Inflammatory-mediated potential confounding effects, including the beneficial gressing . The results of the TEMPO, ELLDOPA (Early reaction. effect of early start therapy per se (see below). versus Late Levodopa) and DATATOP studies would sup- port such an effect. The identification and validation of GDNF preclinical markers for PD would be an important devel- Attempts have also been made to regenerate existing opment. Certain non-motor features, including olfactory dopaminergic neurons. The use of daily intraventricular loss, depression and disorders of rapid-eye-movement or striatal infusions of glial-derived nerve growth factor sleep behaviour, can precede the onset of the motor (GDNF) in MPTP monkeys produced restoration of the symptoms and signs of PD, but none are specific for this nigrostriatal system and improvement in motor function disease118. Nevertheless, it might be possible to enrich an without dyskinesias110. There was a >20% increase in the at-risk population for PD with a combination of these number of hydroxylase-positive nigral neurons, features. Various imaging modalities, including PET with increases in neuronal size, in dopamine levels and a five- 18fluorodopa and single-photon emission computerized fold increase in striatal tyrosine hydroxylase-positive tomography with ligands for the dopamine transporter, fibre density. GDNF delivered by lentivirus transfec- have already found applications in neuroprotection tion one week after MPTP exposure produced similar trials94,95, but the interpretation of the results of these trials remains complex. Further study of these modali- 4 ties as markers of disease progression, even with the Delayed start potentially confounding factor of drug treatment, 3 * is important.

Conclusion 2 **

orsening PD is a progressive multicentric neurodegenerative W 1 disease involving several neurotransmitter systems. Dopamine-replacement therapies have been highly suc-

UP DRS change 0 cessful in improving the motor features of the disease Rasagiline 1 mg but the value of these treatments, particularly !-DOPA, –1 Rasagiline 2 mg

ovement is limited by the development of motor complications. Delayed 2 mg rasagiline There are now several novel therapeutic approaches Impr –2 TABLE 2 0 4 8 14 20 26 32 42 52 emerging for PD, as summerized in . These are Weeks focussed on the non-dopaminergic systems and are Figure 4 | Delayed start TVP-1012 Early Monotherapy for Parkinson’s disease designed to improve motor function without the risk Outpatients (TEMPO) study with rasagiline, 371 subjects. Patients randomised to for motor complications associated with !-DOPA, and early treatment had better clinical outcome at 12 months compared to those early also to improve dyskinesia itself. Disease modification Parkinson’s disease patients on placebo for the first 6 months. UPDRS, Unified Parkinson’s remains the most important goal in PD. Although no Disease Rating Scale *p = 0.05 and **p = 0.01. Adapted, with permission, from REF. 110  drug has yet been proven to be neuroprotective, several (2004) American Medical Association. candidates have shown promise.

852 | OCTOBER 2006 | VOLUME 5 www.nature.com/reviews/drugdisc REVIEWS

1. Braak, H. et al. Staging of brain pathology related to 25. Jeyarasasingam, G., Tompkins, L. & Quik, M. 48. Iravani, M. M. et al. 3,4-methylenedioxymethampheta sporadic Parkinson’s disease. Neurobiol. Aging 24, Stimulation of non-α7 nicotinic receptors partially mine (ecstasy) inhibits dyskinesia expression and 197–211 (2003). protects dopaminergic neurons from 1-methyl-4- normalizes motor activity in 1-methyl-4-phenyl- 2. Rascol, O. et al. A five-year study of the incidence of phenylpyridinium-induced toxicity in culture. 1,2,3,6-tetrahydropyridine-treated primates. dyskinesia in patients with early Parkinson’s disease Neuroscience 109, 275–285 (2002). J. Neurosci. 23, 9107–9115 (2003). who were treated with ropinirole or levodopa. 056 26. Allam, M. F. et al. Smoking and Parkinson’s disease: 49. Chung, K. A., Carlson, N. E. & Nutt, J. G. Short-term Study Group. N. Engl. J. Med. 34, 1484–1491 systematic review of prospective studies. Mov. Disord. paroxetine treatment does not alter the motor 2000). 19, 614–621 (2004). response to levodopa in PD. Neurology 64, 3. Holloway, R. G. et al. Pramipexole vs levodopa as 27. Lucas, G., Bonhomme, N., De Deurwaerdère, P., 1797–1798 (2005). initial treatment for Parkinson disease: a 4-year Le Moal, M. & Spampinato, U. 8-OH-DPAT, a 5-HT1A 50. Meco, G. et al. in L-dopa-induced randomized controlled trial. Arch. Neurol. 61, agonist and , a 5-HT2A/C antagonist, reverse dyskinesias. Clin. Neuropharmacol. 26, 179–181 1044–1053 (2004). -induced catalepsy in rats independently of (2003). 4. Katzenschlager, R. et al. Continuous subcutaneous striatal dopamine release. 131, 51. Parsons, C. G., Danysz, W. & Quack, G. Memantine is apomorphine therapy improves dyskinesias in 57–63 (1997). a clinically well tolerated N-methyl-D-aspartate Parkinson’s disease: a prospective study using 28. De Deurwaerdère, P. & Spampinato, U. Role of (NMDA) receptor antagonist-a review of preclinical single-dose challenges. Mov. Disord. 20, 151–157 serotonin(2A) and serotonin(2B/) receptor subtypes data. 38, 735–767 (999). (2005). in the control of accumbal and striatal dopamine 52. Nash, J. E. et al. The NR2B-selective NMDA receptor 5. Poewe, W. & Luessi, F. Clinical studies with release elicited in vivo by dorsal raphe nucleus electrical antagonist CP-101, 606 exacerbates L-DOPA-induced transdermal rotigotine in early Parkinson’s disease. stimulation. J. Neurochem. 73, 1033–42 (1999). dyskinesia and provides mild potentiation of anti- Neurology 65, S11–S14 (2005). 29. De Deurwaerdère, P., Navailles, S., Berg, K. A., Clarke, parkinsonian effects of L-DOPA in the MPTP-lesioned 6. Newman-Tancredi, A. et al. Differential actions of W. P. & Spampinato, U. Constitutive activity of the marmoset model of Parkinson’s disease. Exp. Neurol. antiparkinson agents at multiple classes of serotonin2C receptor inhibits in vivo dopamine release 188, 471–479 (2004). receptor. II. Agonist and antagonist in the rat striatum and nucleus accumbens. 53. Hadj Tahar, A. et al. Effect of a selective glutamate properties at subtypes of dopamine D(2)-like receptor J. Neurosci. 24, 3235–3241 (2004). antagonist on L-dopa-induced dyskinesias in drug- and α(1)/α(2)-adrenoceptor. J. Pharmacol. Exp. Ther. 30. Nicholson, S. L. & Brotchie, J. M. 5-hydroxytryptamine naive parkinsonian monkeys. Neurobiol. Dis. 15, 303, 805–814 (2002). (5-HT, serotonin) and Parkinson’s disease- 171–176 (2004). 7. Jenner, P. in Principles of Treatment in Parkinson’s opportunities for novel therapeutics to reduce the 54. Counihan, T. J. et al. Expression of N-methyl-D- Disease (eds Schapira, A. H. V. & Olanow, C. W.) problems of levodopa therapy. Eur. J. Neurol. 9 aspartate receptor subunit mRNA in the human brain: 33–359 (Elsevier, Philadelphia, 2005). (Suppl. 3), 1–6 (2002). mesencephalic dopaminergic neurons. J. Comp. 8. Fahn, S. Does levodopa slow or hasten the rate of 31. Frechilla, D. et al. Serotonin 5-HT(1A) receptor Neurol. 390, 91–101 (1998). progression of Parkinson’s disease? J. Neurol. 252 expression is selectively enhanced in the striosomal 55. Feng, B. et al. Structure-activity analysis of a novel (Suppl. 4), iv37–iv42 (2005). compartment of chronic parkinsonian monkeys. NR2C/NR2D-preferring NMDA receptor antagonist: 9. Olanow, W., Schapira, A. H. & Rascol, O. Continuous Synapse 39, 288–296 (2001). 1-(phenanthrene-2-carbonyl) -2,3- dopamine-receptor stimulation in early Parkinson’s 32. Numan, S. et al. Increased expression of 5HT2 dicarboxylic acid. Br. J. Pharmacol. 141, 508–516 disease. Trends Neurosci. 23, S117–S126 (2000). receptor mRNA in rat striatum following 6-OHDA (2004). 10. Calon, F. et al. Dopamine-receptor stimulation: lesions of the adult nigrostriatal pathway. Brain Res. 56. Picconi, B. et al. Loss of bidirectional striatal synaptic biobehavioral and biochemical consequences. Trends Mol. Brain Res. 29, 391–396 (1995). plasticity in L-DOPA-induced dyskinesia. Nature Neurosci. 23, S92–S100 (2000). 33. Fox, S. H. & Brotchie, J. M. 5-HT2C receptor binding Neurosci. 6, 501–506 (2003). 11. Brooks, D. J. & Sagar, H. Entacapone is beneficial in is increased in the substantia nigra pars reticulata in 57. Marino, M. J. et al. Allosteric modulation of group III both fluctuating and non-fluctuating patients with Parkinson’s disease. Mov. Disord. 15, 1064–1069 metabotropic 4: a potential Parkinson’s disease: a randomised, placebo (2000). approach to Parkinson’s disease treatment. Proc. Natl controlled, double blind, six month study. J. Neurol. 34. Bibbiani, F., Oh, J. D. & Chase, T. N. Serotonin 5-HT1A Acad. Sci. USA 100, 13668–13673 (2003). Neurosurg. Psychiatry 74, 1071–1079 (2003). agonist improves motor complications in rodent and 58. O’Neill, M. J. et al. Neurotrophic actions of the novel 12. Mercuri, N. B. & Bernardi, G. The ‘magic’ of L-dopa: primate parkinsonian models. Neurology. 57, AMPA receptor potentiator, LY404187, in rodent why is it the gold standard Parkinson’s disease 1829–1834 (2001). models of Parkinson’s disease. Eur. J. Pharmacol. therapy? Trends Pharmacol. Sci. 26, 341–344 35. Chassain, C. & Durif, F. M. Modulation of serotonin 486, 163–174 (2004). (2005). receptors in MPTP-lesioned monkeys. Mov. Disord. 59. More, J. C. et al. Characterisation of UBP296: a novel, 13. Perry, E. K. et al. Cholinergic nicotinic and muscarinic 19, 412 (2004). potent and selective kainate receptor antagonist. receptors in dementia of Alzheimer, Parkinson and 36. Olanow, C. W. et al. Multicenter, open-label, trial of Neuropharmacology 47, 46–64 (2004). Lewy body types. J. Neural. Transm. Park. Dis. sarizotan in Parkinson disease patients with levodopa- 60. Calon, F. et al. Levodopa or D2 agonist induced Dement. Sect. 2, 149–158 (1990). induced dyskinesias (the SPLENDID Study). Clin. dyskinesia in MPTP monkeys: correlation with changes

14. Asahina, M. et al. Hypersensitivity of cortical Neuropharmacol. 27, 58–62 (2004). in dopamine and GABAA receptors in the muscarinic receptors in Parkinson’s disease 37. Bara-Jimenez, W. et al. Effects of serotonin 5-HT1A striatopallidal complex. Brain Res. 680, 43–52 demonstrated by PET. Acta Neurol. Scand. 91, agonist in advanced Parkinson’s disease. Mov. Disord. (1995). 437–443 (1995). 20, 932–936 (2005). 61. Calon, F. et al. Changes of GABA receptors and 15. Quik, M. Smoking, nicotine and Parkinson’s disease. 38. Bonifati, V. et al. Buspirone in levodopa-induced dopamine turnover in the postmortem brains of Trends Neurosci. 27, 561–568 (2004). dyskinesias. Clin. Neuropharmacol. 17, 73–82 (1994). parkinsonians with levodopa-induced motor 16. Rodriguez-Puertas, R., Pazos, A. & Pascual, J. 39. Oh, J. D., Bibbiani, F. & Chase, T. N. Quetiapine complications. Mov. Disord. 18, 241–253 (2003). Cholinergic markers in degenerative parkinsonism: attenuates levodopa-induced motor complications in 62. Xu, K., Bastia, E. & Schwarzschild, M. Therapeutic autoradiographic demonstration of high-affinity rodent and primate parkinsonian models. Exp. Neurol. potential of adenosine A(2A) receptor antagonists in choline uptake carrier hyperactivity. Brain Res. 636, 177, 557–564 (2002). Parkinson’s disease. Pharmacol. Ther. 105, 267–310 327–332 (1994). 40. Fox, S. H., Moser, B. & Brotchie, J. M. Behavioral (2005). 17. Katzenschlager, R. et al. Anticholinergics for effects of 5-HT2C receptor antagonism in the 63. Jenner, P. Istradefylline, a novel adenosine A2A receptor symptomatic management of Parkinson’s disease. substantia nigra zona reticulata of the antagonist, for the treatment of Parkinson’s disease. Cochrane Database Syst Rev. CD003735 (2003). 6-hydroxydopamine-lesioned rat model of Parkinson’s Expert Opin. Investig. Drugs. 14, 729–738 (2005). 18. Leroi, I. et al. Randomized placebo-controlled trial of disease. Exp. Neurol. 151, 35–49 (1998). 64. Fink, J S. et al. Molecular cloning of the rat A2 donepezil in cognitive impairment in Parkinson’s 41. Fox, S. H. & Brotchie, J. M. 5-HT(2C) receptor : selective co-expression with D2 disease. Int. J. Geriatr. Psychiatry 19, 1–8 (2004). antagonists enhance the behavioural response to dopamine receptors in rat striatum. Brain Res. Mol. 19. Aarsland, D. et al. Donepezil for cognitive impairment dopamine D(1) receptor agonists in the Brain Res. 14, 186–195 (1992). in Parkinson’s disease: a randomised controlled study. 6-hydroxydopamine-lesioned rat. Eur. J. Pharmacol. 65. Grondin, R.et al. Antiparkinsonian effect of a new J. Neurol. Neurosurg. Psychiatry 72, 708–712 398, 59–64 (2000). selective adenosine A2A receptor antagonist in (2002). 42. Gomez-Mancilla, B. & Bedard, P. J. Effect of MPTP-treated monkeys. Neurology 52, 1673–1637 20. Emre, M. et al. Rivastigmine for dementia associated nondopaminergic drugs on L-dopa-induced dyskinesias (1999). with Parkinson’s disease. N. Engl. J. Med. 351, in MPTP-treated monkeys. Clin. Neuropharmacol. 16, 66. Mori, A. & Shindou, T. Modulation of GABAergic 2509–2518 (2004). 418–427 (1993). transmission in the striatopallidal system by 21. Werber, E. A. & Rabey, J. M. The beneficial effect of 43. Ashby, C. R. Jr. & Wang, R. Y. Pharmacological actions adenosine A2A receptors: a potential mechanism for cholinesterase inhibitors on patients suffering from of the atypical drug clozapine: a review. the antiparkinsonian effects of A2A antagonists. Parkinson’s disease and dementia. J. Neural Transm. Synapse 24, 349–394 (1996). Neurology 61, S44–S48 (2003). 108, 1319–1325 (2001). 44. Bymaster, F. P. et al. Radioreceptor binding profile of 67. Bara-Jimenez, W. et al. Adenosine A(2A) receptor 22. Fabbrini, G. et al. Donepezil in the treatment of the . antagonist treatment of Parkinson’s disease. hallucinations and delusions in Parkinson’s disease. Neuropsychopharmacology 14, 87–96 (1996). Neurology 61, 293–296 (2003). Neurol. Sci. 23, 41–43 (2002). 45. Grondin, R. et al. D1 receptor blockade improves 68. Hauser, R. A., Hubble, J. P. & Truong, D. D. 23. Le Novere, N., Zoli, M. & Changeux, J. P. Neuronal L-dopa-induced dyskinesia but worsens parkinsonism Randomized trial of the adenosine A(2A) receptor nicotinic receptor α6 subunit mRNA is selectively in MPTP monkeys. Neurology 52, 771–776 (1999). antagonist istradefylline in advanced PD. Neurology concentrated in nuclei of the 46. Durif, F. et al. Clozapine improves dyskinesias in 61, 297–303 (2003). rat brain. Eur. J. Neurosci. 8, 2428–2439 Parkinson disease: a double-blind, placebo-controlled 69. Ascherio, A. & Chen, H. Caffeinated clues from (1996). study. Neurology 62, 381–388 (2004). epidemiology of Parkinson’s disease. Neurology 61, 24. Costa, G., Abin-Carriquiry, J. A. & Dajas, F. Nicotine 47. Katzenschlager, R. et al. Low dose quetiapine for drug S51–S54 (2003). prevents striatal dopamine loss produced by induced dyskinesias in Parkinson’s disease: a double 70. Chen, J. F. et al. Neuroprotection by caffeine and 6-hydroxydopamine lesion in the substantia nigra. blind cross over study. J. Neurol. Neurosurg. A(2A) adenosine receptor inactivation in a model of Brain Res. 888, 336–342 (2001). Psychiatry 75, 295–297 (2004). Parkinson’s disease. J. Neurosci. 21, RC143 (2001).

NATURE REVIEWS | DRUG DISCOVERY VOLUME 5 | OCTOBER 2006 | 853 REVIEWS

71. Fredduzzi, S. et al. Persistent behavioral sensitization 89. Hunot, S. & Flavell, R. A. Apoptosis. Death of a 108. Maruyama, W., Takahashi, T., Youdim, M. & Naoi, M. to chronic L-DOPA requires A2A adenosine receptors. monopoly? Science 292, 865–866 (2001). The anti-Parkinson drug, rasagiline, prevents J. Neurosci. 22, 1054–1062 (2002). 90. Chen, H. et al. Nonsteroidal anti-inflammatory drugs apoptotic DNA damage induced by peroxynitrite in 72. Bibbiani, F. et al. A2A antagonist prevents dopamine and the risk of Parkinson disease. Arch. Neurol. 60, human dopaminergic neuroblastoma SH-SY5Y cells. agonist-induced motor complications in animal models 1059–1064 (2003). J. Neural Transm. 109, 467–481 (2002). of Parkinson’s disease. Exp. Neurol. 184, 285–294 91. Schapira, A. H. Disease modification in Parkinson’s 109. Jenner, P. Preclinical evidence for neuroprotection (2003). disease. Lancet Neurol. 3, 362–368 (2004). with monoamine oxidase-B inhibitors in Parkinson’s 73. Fuxe, K. et al. Receptor heteromerization in adenosine 92. Iravani, M. M., Haddon, C. O., Cooper, J. M., Jenner, P. disease. Neurology 63 (Suppl. 2), S13–S22 (2004). A2A receptor signaling: relevance for striatal function & Schapira, A. H. Pramipexole protects against MPTP 110. Parkinson Study Group. A controlled, randomized, and Parkinson’s disease. Neurology 61, S19–S23 toxicity in non-human primates. J. Neurochem. 96, delayed-start study of rasagiline in early Parkinson (2003). 1315–1321 (2006). disease. Arch. Neurol. 61, 561–566 (2004). 74. Kachroo, A. et al. Interactions between metabotropic 93. Gu, M. et al. Pramipexole protects against apoptotic 111. Grondin, R. et al. Chronic, controlled GDNF infusion glutamate 5 and adenosine A2A receptors in cell death by non-dopaminergic mechanisms. promotes structural and functional recovery in advanced normal and parkinsonian mice. J. Neurosci. 25, J. Neurochem. 91, 1075–1081 (2004). parkinsonian monkeys. Brain 125, 2191–2201 (2002). 10414–10419 (2005). 94. Whone, A. L. et al. Slower progression of Parkinson’s 112. Kordower, J. H. et al. prevented 75. Coccurello, R., Breysse, N. & Amalric, M. disease with ropinirole versus levodopa: The REAL- by lentiviral vector delivery of GDNF in primate

Simultaneous blockade of adenosine A2A and PET study. Ann. Neurol. 54, 93–101 (2003). models of Parkinson’s disease. Science 290, metabotropic glutamate mGlu5 receptors increase 95. Parkinson Study Group. Dopamine transporter brain 767–773 (2000). their efficacy in reversing Parkinsonian deficits in rats. imaging to assess the effects of pramipexole vs 113. Kordower, J. H. et al. Clinicopathological findings Neuropsychopharmacology 29, 1451–161 (2004). levodopa on Parkinson disease progression. JAMA following intraventricular glial-derived neurotrophic 76. Bezard, E., Brotchie, J. M. & Gross, C. E. 287, 1653–1661 (2002). factor treatment in a patient with Parkinson’s disease. Pathophysiology of levodopa-induced dyskinesia: 96. Schapira, A. H. & Olanow, C. W. Neuroprotection in Ann. Neurol. 46, 419–424 (1999). potential for new therapies. Nature Rev. Neurosci. 2, Parkinson disease: mysteries, myths, and 114. Nutt, J. G. et al. Randomized, double-blind trial of 577–588 (2001). misconceptions. JAMA 291, 358–364 (2004). glial cell line-derived neurotrophic factor (GDNF) in 77. Calon, F. et al. Increase of preproenkephalin mRNA 97. Mytilineou, C. & Cohen, G. Deprenyl protects PD. Neurology 60, 69–73 (2003). levels in the putamen of Parkinson disease patients dopamine neurons from the neurotoxic effect of 115. Gill, S. S. et al. Direct brain infusion of glial cell line- with levodopa-induced dyskinesias. J. Neuropathol. 1-methyl-4-phenylpyridinium ion. J. Neurochem. 45, derived neurotrophic factor in Parkinson disease. Exp. Neurol. 61, 186–196 (2002). 1951–1953 (1985). Nature Med. 9, 589–595 (2003). 78. Henry, B. et al. µ- and δ- antagonists 98. Tatton, W. G. & Greenwood, C. E. Rescue of dying 116. Lang, A. E. et al. Randomized controlled trial of reduce levodopa-induced dyskinesia in the MPTP- neurons: a new action for deprenyl in MPTP intraputamenal glial cell line-derived neurotrophic lesioned primate model of Parkinson’s disease. Exp. parkinsonism. J. Neurosci. Res. 30, 666–672 (1991). factor infusion in Parkinson disease. Ann. Neurol. 59, Neurol. 171, 139–146 (2001). 99. Chiueh, C. C., Huang, S. J. & Murphy, D. L. Enhanced 459–466 (2006). 79. Samadi, P., Gregoire, L. & Bedard, P. J. Opioid hydroxyl radical generation by 2′-methyl analog of Barker, R. A. Continuing trials of GDNF in Parkinson’s antagonists increase the dyskinetic response to MPTP: suppression by clorgyline and deprenyl. disease. Lancet Neurol. 5, 285–286 (2006). dopaminergic agents in parkinsonian monkeys: Synapse 11, 346–348 (1992). 117. Schapira, A. H. & Obeso, J. Timing of treatment interaction between dopamine and opioid systems. 100. Wu RM, Chiueh CC, Pert A, Murphy DL. Apparent initiation in Parkinson’s disease: a need for Neuropharmacology 45, 954–963 (2003). antioxidant effect of L-deprenyl on hydroxyl radical reappraisal? Ann. Neurol. 59, 559–562 (2006). 80. Fox, S. et al. Non-subtype-selective opioid receptor formation and nigral injury elicited by MPP+ in vivo. 118. Chaudhuri, K. R., Healy, D. G. & Schapira, A. H. Non- antagonism in treatment of levodopa-induced motor Eur. J. Pharmacol. 243, 241–247 (1993). motor symptoms of Parkinson’s disease: diagnosis and complications in Parkinson’s disease. Mov. Disord. 19, 102. Wu, R. M., Chen, R. C. & Chiueh, C. C. Effect of management Lancet Neurol. 5, 235–245 (2006). 554–560 (2004). MAO-B inhibitors on MPP+ toxicity in vivo. Ann. NY 119. Lang, A. E. & Obeso, J. A. Challenges in Parkinson’s 81. Rascol, O. et al. Naltrexone, an opiate antagonist, fails Acad. Sci. 899, 255–261 (2000). disease: restoration of the nigrostriatal dopamine system to modify motor symptoms in patients with Parkinson’s 103. Tatton, W. G., Ju, W. Y., Holland, D. P., Tai, C. & Kwan, is not enough. Lancet Neurol. 3, 309–316 (2004). disease. Mov. Disord. 9, 437–440 (1994). M. (–)-Deprenyl reduces PC12 cell apoptosis by 82. Hollt, V. Opioid peptide processing and receptor inducing new protein synthesis. J. Neurochem. 63, Competing interests statement selectivity. Annu. Rev. Pharmacol. Toxicol. 26, 59–77 1572–1575 (1994). The authors declare competing financial interests: see Web (1986). 104. Wadia, J. S. et al. Mitochondrial membrane potential version for details. 83. Sadee, W., Wang, D. & Bilsky, E. J. Basal opioid and nuclear changes in apoptosis caused by serum receptor activity, neutral antagonists, and therapeutic and nerve growth factor withdrawal: time course and DATABASES opportunities. Life Sci. 76, 1427–1437 (2005). modification by (–)-deprenyl. J. Neurosci. 18, The following terms in this article are linked online to: 84. Simonds, W. F. The molecular basis of opioid receptor 932–947 (1998). Entrez Gene: function. Endocr Rev. 9, 200–212 (1988). 105. Parkinson Disease Study Group Effects of tocopherol http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene 85. Mansour, A. et al. Opioid-receptor mRNA expression and deprenyl on the progression of disability in early 5-HT1a | 5-HT1B | 5-HT2A | 5-HT2C | α-synuclein | A2AR | DJ1 | in the rat CNS: anatomical and functional implications. Parkinson’s disease. N. Engl. J. Med. 328, 176–183 GDNF | HTRA2 | LRRK2 | mGlu4 | mGlu5 | NR2B | NURR1 | Trends Neurosci. 18, 22–29 (1995). (1993). Parkin | Preproenkephalin-A | PINK1 | UCHL1 86. Schapira, A. H. Etiology of Parkinson’s disease. 106. Akao, Y. et al. An anti-Parkinson’s disease drug, OMIM: Neurology 66, S10–S23 (2006). N-propargyl-1(R)-aminoindan (rasagiline), enhances http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM 87. Dawson, T. M. & Dawson, V. L. Molecular pathways of expression of anti-apoptotic bcl-2 in human Parkinson’s disease neurodegeneration in Parkinson’s disease. Science. dopaminergic SH-SY5Y cells. Neurosci. Lett. 326, 302, 819–822 (2003). 105–108 (2002). FURTHER INFORMATION 88. Shults, C. W. et al. Effects of coenzyme Q10 in early 107. Am, O. B., Amit, T. & Youdim, M. B. Contrasting Clinical Trials Homepage: Parkinson disease: evidence of slowing of the neuroprotective and neurotoxic actions of respective http://www.cancer.gov/clinicaltrials functional decline. Arch. Neurol. 59, 1541–1550 metabolites of anti-Parkinson drugs rasagiline and Access to this links box is available online. (2002). selegiline. Neurosci. Lett. 355, 169–172 (2004).

854 | OCTOBER 2006 | VOLUME 5 www.nature.com/reviews/drugdisc