Modulations of the amide function of the preferential dopamine D3 agonist (R,R)-S32504: Improvements of affinity and selectivity for D3 versus D2 receptors

Article abstract of DOI:10.1016/j.bmcl.2009.03.015

Starting with the preferential dopamine (DA) D₃ agonist S32504, we prepared two series of derivatives of the general formula I-A and I-B, in an effort to improve both potency and selectivity. For the first set of derivatives, where the primary

Full text of DOI:10.1016/j.bmcl.2009.03.015

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2133–2138
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Modulations of the amide function of the preferential dopamine D3
agonist (R,R)-S32504: Improvements of affinity and selectivity for D3
versus D2 receptors
a
b
b
b
Jean-Louis Peglion a, , Christophe Poitevin , Clotilde Mannoury La Cour , Delphine Dupuis , Mark J. Millan
*
a Institut de Recherches Servier, Chemistry B Departement, 11 rue des Moulineaux, 92150 Suresnes, France
b Institut de Recherches Servier, Psychopharmacology Departement, 125 chemin de Ronde, 78290 Croissy, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Starting with the preferential dopamine (DA) D3 agonist S32504, we prepared two series of derivatives of
the general formula I-A and I-B, in an effort to improve both potency and selectivity. For the first set of
derivatives, where the primary amide function of S32504 was replaced by either secondary and tertiary
amide or ester, acid, nitrile and ketone, no improvement was obtained. Conversely, when the primary
amide function was integrated in a lactam ring, an enhancement of affinity and selectivity was attained
for the five-membered ring lactam but also for its five-membered ring lactone analogue.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 2 February 2009
Revised 2 March 2009
Accepted 3 March 2009
Available online 9 March 2009
Keywords:
Dopamine
D3 receptors
Parkinson
Ortholithiation
Cyclization
Dopaminergic mechanisms are broadly involved in the control
of diverse physiological functions and their perturbation is impli-
cated in the etiology of debilitating central nervous system disor-
ders like schizophrenia, depression and Parkinson’s disease (PD).
Consequently, dopaminergic ligands are of considerable interest
as therapeutic agents and elucidation of the functional roles of
the five individual classes of DA receptor (D1–D5) is of great
importance.1
beneficial D3 receptor-mediated neuroprotective effects on dopa-
minergic neurones have been documented in cellular models, in
rodents and in man.11,12
In our continuing effort to identify ligands interacting preferen-
tially with D3 receptors, we have already characterized
S14297(1)13–15 and a series of tetracyclic analogues16 of S14297 as
potent hD3 antagonists.17 More recently we discovered S32504
(R,R)-2 which behaves as D3 receptor agonist18 (Fig. 2) and displays
robust antiparkinsonian, neuroprotective and antidepressive activ-
ity in vivo in comparison to other agents such as ropinirole.19,20
In order to further increase the potency of S32504 and propose
improved treatments of dopaminergic-related diseases (and more
particularly PD) we decided to extensively modulate its amide
function. To this end we elected to prepare compounds of general
formula I-A and I-B (Fig. 2) to explore the influence of both minor
modulations (amide bioisosteres, I-A) and more important trans-
formations (rigidified analogues, I-B) of this functional group, upon
potency at hD3 receptors and selectivity versus hD2 receptors.
Compounds of general formula I-A were prepared from triflate
Since its discovery by Sokoloff and Schwartz2 in the early 1990s,
dopamine D3 receptors, which display a pattern of ligand recogni-
tion and intracellular coupling similar but subtly different to clo-
sely-related D2 receptors, have attracted much attention.3–5 D3
receptors share the importance of D2 receptors in the control of
mood, cognition, emotion, reward and motor function but, inter-
estingly, their role is distinct. For example, D3 but not D2 receptor
antagonists improve cognition and reduce relapse of drugs of
abuse, while blockade of D2 receptors is the core mechanism for
controlling positive symptoms of schizophrenia.6,7
The respective significance of D3 versus D2 receptor stimulation
to the control of symptoms in PD remains unclear, but preferential
D3 versus D2 receptor agonists like ropinirole and pramipexole
(Fig. 1), and the partial D3 (and D2) receptor agonist piribedil8 are
widely used as antiparkinson agents.9,10 Of particular significance,
3
21 (Scheme 1). By action of zinc cyanide in presence of a catalytic
amount of palladium tetrakis triphenylphosphine in DMF,22 3 was
transformed in 4 in moderate yield (36%), which was further mod-
ified in ( )S32504 (( )2) by treatment with potassium hydroxide in
ethanol at reflux (80% yield). Ester 5 was obtained from 2, in 65%
yield, by action of 1,1-dimethoxy N,N-dimethylformamide (DMF–
DMA) and then submitting the derivative obtained to sodium
* Corresponding author. Tel.: +33 1 55 72 22 13; fax: +33 1 55 72 24 30.
E-mail address: jean-louis.peglion@fr.netgrs.com (J.-L. Peglion).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.015

2134
J.-L. Peglion et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2133–2138
O
O
O
H
N
N
S
HN
N
N
N
H2N
N
N
Ropinirole
Pramipexole
Piribedil
Figure 1. Structure of the preferential D3 versus D2 receptor agonists, ropinirole and pramipexole, and of the non-ergot D3/D2 partial agonist, piribedil.
O
O
H
N
O
N
H
H2N
S 14297
1
S 32504 (R,R)-2
O
H
O
O
A = CONR1R2
COOR
N
A
N
X
H
CN
COCH3
X = O, NR1
n = 0, 1
n
CN4H
I-A
I-B
Figure 2. Structure of preferential antagonists and agonists at hD3 versus hD2 receptors.
methylate according to Anelli et al.23 and Ono et al.24 Finally, acid
derivative 6 was obtained by basic hydrolysis of 5 using sodium
hydroxide.
quently subsequent modulations were undertaken only on the cor-
responding R,R isomer (see Fig. 3).
On the structure of 2, both position 8 and 10 of the hexahydro-
2H-naphtho[1,2-b] [1,4] oxazine backbone were available for
anchoring an extra cycle incorporating the amide function. Consid-
ering the X-ray structure of S32504 we decided to attach this
fourth ring in position 8. The resulting rigidified or tetracyclic ana-
logues I-B (15–20) of (R,R)-S32504 were prepared from (R,R)-6.
The lactam derivative 15 was synthetized as depicted in Scheme
3. To prepare the cyano derivative 22, we first synthetized the N,N-
diethylamide derivative 21 (76% yield using the same experimental
conditions as for the preparation of 10–14) to be able to perform an
ortho directed lithiation at temperature below ꢀ70 °C, followed by
Excess butylvinyl ether in presence of 3 mol % of 1,3-
bis(diphenylphosphino)propane (Dppp) and 2.5 mol % of palla-
dium acetate allowed the transformation of triflate 3 in methyl ke-
tone 7, in 66% yield, according to Cabri et al.25 Tetrazole 8 was
prepared in 60% yield by a 1,3-dipolar cycloaddition of the cyano
derivative 4 with trimethyltin azide in toluene followed by treat-
ment with HCl gas, according to Duncia et al.26 Finally, action of
gaseous hydrochloric acid in ethanol on this same cyano derivative
led to ethyl ester 9.
From an analysis of the data reported in Table 1, it can be seen
that the amide function present in 2 is the upmost functionality for
affinity at D3 receptors and selectivity versus D2 receptors. Cyanide
as in 4 and acetyl as in 7, although conferring some selectivity, did
not allowed for an improvement in affinity at D3 receptors but led
to derivatives of comparable potency than the reference compound
ropinirole. Others functions examined, such as esters in 5 and 9,
carboxylic acid in 6 and tetrazole in 8, were detrimental for affinity
since these compounds are almost devoid of activity. Compound 2,
the most potent and selective of this first set of compounds was re-
solved by chiral HPLC.27 (+)228 retained the affinity of the racemate
and was two fold more selective, whereas (ꢀ)2 was practically de-
void of activity.
Next, we decided to examine the role of nitrogen substituants
on the amide function and to this end synthetized amides 10–14
by coupling the corresponding amines with 6, in the presence of
TBTU (O-(benzotriazol-1-yl) N,N,N0,N0-tetramethyluronium tetra-
fluoroborate) and (Et)3N as depicted in Scheme 2 (yields 38–
56%). Mono substitutions as in 10, 12, 13, 14 or di-substitution as
in 11 dramatically decreased affinity for hD3 receptors while hav-
ing little impact on hD2 affinities.
At this point and before further modifications of the amide part
of S32504 it was deemed worthwhile to determine the absolute
configuration of asymmetric carbons at ring junctions, since we
did not envisaged to change this part of the molecule. For this pur-
pose, we performed X-ray analysis on the methanesulfonate salt of
the eutomer (+)2, which revealed a R,R configuration.29 Conse-
Table 1
Affinities of compounds 2–14 (I-A) at hD3 and hD2 dopamine receptors as determined
by displacement of [3H]-spiperone
Affinity (pKi)a
Selectivityb
hD3
hD2L
Ki(hD2)/Ki(hD3)
( )2
(+)(R,R)2
7.71 0.06
7.83 0.15
5.61 0.04
7.40 0.05
<5
5.48 0.12
7.02 0.01
<5
6.12 0.09
5.92 0.10
<5
6.10 0.18
<6
40
80
(ꢀ)(S,S)2
4
5
6
7
20
32
<5
5.51 0.11
<5
8
9
10
11
12
13
14
5.58 0.13
<5
<5
<5
<5
N.T.
5.10 0.08
5.73 0.06
5.34 0.01
5.47 0.07
<5
5.56 0.12
7.44 0.06
Ropinirole
6.16 0.06
20
a
pKi values are derived from at least two independent determinations each
performed in duplicate. Experiments were undertaken on membranes of CHO cells
stably transfected with cloned hD3 or hD2 receptors (hD2L, long isoform). [3H]-
spiperone was used as the radioligand.
b
Selectivity is expressed by the ratio of Ki (nM) values at hD2 versus hD3
receptors.

J.-L. Peglion et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2133–2138
2135
pound 16 was achieved via treatment at reflux with HCl 6N. The
overall yield for the two last steps was 12%.32 Subsequently, reac-
tion of 16 with methylamine at 120 °C led to 17 with a yield of 25%.
Lactone 18 and lactam 19 were prepared as depicted in Scheme
5 starting from amide 21. The six-membered ring lactone 18 was
obtained like the five-membered ring lactone 16 but using the
lithio protected 2-bromoethanol as the electrophile instead of
DMF. The overall yield for these two steps was only 10% but with-
out any evidence of the other potential regioisomer. From alcohol
25, lactam 19 was obtained through the sequence: chloride 27
(yield 75%), nitrile 28 (yield 96%), amine 29 (yield 30%), and lactam
(yield 49%).33
Unfortunately, N,N-diethylamide 21 was not suitable for prepa-
ration of lactam 20, which was synthetized according to Scheme 6,
adapted from Fisher et al.34 We anticipated that Weinreb’s amide,
easily obtained in 2 steps from 6 and with a yield of 61%, could
serve as the ideal group for: (1) ortho directed metalation leading
to 32 after reaction with methyl iodide, which (2) underwent reg-
iospecific dilithiation on nitrogen and methyl and (3) in presence
of DMF, allowed for the preparation of the N-protected methoxy
lactam 33, via a cyclization under mildly acidic conditions with
an acceptable yield of 25% for the three steps. Finally deprotection
was achieved with a moderate yield (31%) by use of titanium tri-
chloride in ethanol at reflux.
Several conclusions can be drawn from the affinities and selec-
tivity’s obtained with this second set of derivatives (Table 2). For
rigidified derivatives, affinities at hD3 and selectivity’s versus hD2
sites were markedly increased in comparison to derivatives pre-
sented in Table 1. Compound (R,R)-15 shows the highest hD3
receptor affinity amongst all compounds prepared in this study,
Figure 3. X-ray structure of the methanesulfonate salt of (+)2.
an electrophilic cyanation using phenylcyanate (PhOCN), as de-
scribed by Sato.30 Probably due to steric hindrance we were fortu-
nate to isolate only 22 in 78% yield and without any evidence of
cyanation in position 10. Then, 22 was submitted to a catalytic
hydrogenation and 23, which was isolated in 31% yield, was cy-
clized to yield compound 15 in presence of t-BuLi at ꢀ78 °C (99%
yield).31
Lactone 16 and lactam 17 were prepared as depicted in Scheme
4 following a synthetic pathway similar to the one used for prepa-
ration of 15. Intermediate 21 was lithiated, as previously, by sec-
butyllithium in presence of N,N,N0,N0-tetramethylethylenediamine
(TMEDA) and the resulting lithoamide, treated by DMF in THF,
led to 24. Here again and despite a comparatively modest yield
(47%) we only isolated the desired ortho isomer. Reduction with
sodium borohydride afforded alcohol 25 and cyclization to com-
O
TfO
N
O
O
e
N
3
4
9
H2N
O
O
2
a
b
N
c
O
N
N
7
O
O
N
O
f
5
6
N
O
N
g
N
N
d
N
H
O
O
O
O
N
8
N
O
HO
Scheme 1. Structure and preparation of compounds 2–9. Reagents: (a) Zn(CN)2, Pd[P(Ph)3]4, DMF, 36%; (b) KOH, EtOH, reflux, 80%; (c) (i) DMF–DMA, MeOH, (ii) MeONa,
MeOH, THF, 65%; (d) NaOH, MeOH, reflux, 80%; (e) butylvinyl ether, NEt3, Dppp, Pd(OAc)2, DMF, HCl, heat 80 °C, 66%; (f) Me3SnN3, toluene, reflux, 60%; (g) EtOH, HClg,
phosphate buffer, 85%.
10 R1 = CH3, R2 = H ; 44%
O
O
O
O
11 R1, R2 = CH3 ; 47%
R1
N
N
HO
12 R1 = nBu, R2 = H ; 56%
a
N
R1R2NH/NEt3
TBTU/CH2Cl2
1
, R2 = H ; 38%
, R2 = H ; 43%
R2
O
13 R =
6
10-14
1
O
14 R =
Scheme 2. Preparation of compounds (10–14).

2136
J.-L. Peglion et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2133–2138
O
OH
O
N
O
N
H
H
H
N
H
N
H
N
H
O
O
O
N
a
b
21
22
(R,R)-6
O
N
O
O
H
H
N
H
N
H
d
c
O
HN
H2N
23
(R,R)-15
Scheme 3. Preparation of compound 15. Reagents: (a) NH(Et)2, NEt3, TBTU, CH2Cl2, 76%; (b) (i) s-BuLi, TMEDA, THF, (ii) PhOCN, 78%; (c) H2, Raney Ni, MeOH, 31%; (d) tBuLi,
THF, 99%.
O
N
O
N
H
N
O
H
H
H
N
H
N
H
N
H
O
a
b
O
O
O
HO
24
25
21
O
O
O
O
H
H
N
H
N
H
c
d
O
N
(R,R)-17
(R,R)-16
Scheme 4. Preparation of compounds 16 and 17. Reagents: (a) (i) s-BuLi, TMEDA, THF, (ii) DMF, 47%; (b) NaBH4, MeOH, 34%; (c) HCl 6 N, reflux, 35%; (d) MeNH2, 120 °C, 25%.
O
O
O
N
O
N
H
H
H
N
H
N
H
N
H
b
O
O
a
O
HO
26
(R,R)-18
21
c, d
O
H
O
O
N
O
N
H
H
N
H
N
H
g
N
h
HN
O
O
H
X
H2N
(R,R)-19
29
25 X = OH
27 X = Cl
28 X = CN
e
f
Scheme 5. Preparation of compounds 18 and 19. Reagents: (a) (i) s-BuLi, TMEDA, THF, (ii) LiOEtBr, 17%; (b) HCl 6 N, reflux, 57%; (c) (i) s-BuLi, TMEDA, THF, (ii) DMF, 47%; (d)
NaBH4, MeOH, 34%; (e) SOCl2, 75%; (f) (Bu)4N+NCꢀ (TBACN), THF, 96%; (g) H2, Raney Ni, MeOH, 30%;, (h) t-BuLi, THF, ꢀ78 °C, 49%.
and the amelioration of selectivity in comparison to (R,R)-2 was
2.5-fold. Moreover, the optical opposite (S,S)-15 manifested only
weak affinity at hD3 receptors. Surprisingly, the second best affin-
ity at hD3 receptors and the most pronounced selectivity was re-
vealed by the five-membered ring lactone (R,R)-16, which can be
viewed as a rigidified analogue of the inactive ester 5 and acid 6.
In addition, the six-membered ring lactone 18 also interacted with
hD3 receptors, though displaying a modest pKi of 6.9. On the other
hand, the size of the extracycle seems to play a critical role for
selectivity since the six-membered lactam 19 displayed 10-fold
lower selectivity than the five-membered derivate (R,R)-15, a char-
acteristic shared by lactone 18 which presented a selectivity
twenty times inferior to lactone 16 due to a decrease of hD3 affinity
without modification of affinity at hD2 sites. Affinity also seems
dependent on the size of the cycle, since the six-membered deriv-
atives 18 and 19 were about 10 times less potent than their five-
membered analogues (R,R)-15 and 16. A further point to be raised
is that the presence of a double bond in the six-membered cycle in-

J.-L. Peglion et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2133–2138
2137
O
O
O
O
HN
O
Cl
H
H
H
N
H
N
H
N
H
b
a
HO
O
O
(R,R)-6
31
30
O
O
O
O
O
O
HN
H
H
H
N
H
N
H
O
N
H
c
d
e
HN
N
O
32
(R,R)-20
33
Scheme 6. Preparation of compound 20. Reagents and conditions: (a) SOCl2, DMF, toluene, 97%; (b) MeONH2, K2CO3, AcOEt, H2O, 63%; (c) (i) s-BuLi, TMEDA, THF, (ii) CH3I; (d)
(i) s-BuLi, DMF, THF, (ii) HCl, THF, 25% on two steps; (e) TiCl3, EtOH, reflux, 31%.
Table 2
disubstituted amide derivatives were devoid of activity. On the
other hand, transformation of this primary amide function in lac-
tam (R,R)-15 or lactone (R,R)-16 improved both potency and
selectivity, providing that the additional ring was five-membered
one. A five-membered ring is also requisite for maximal efficacy
since the six-membered lactam (R,R)-19 was less active. Since
(R,R)-15 behaved as a potent D3 agonist in vitro with marked
Affinities of compounds 15–20 (I-B) at hD3 and hD2 dopamine receptors as
determined by displacement of [3H]-spiperone
Affinity (pKi)a
Selectivityb
hD3
hD2L
Ki(hD2)/Ki(hD3)
(R,R)-15
(S,S)-15
(R,R)-16
(R,R)-17
(R,R)-18
(R,R)-19
(R,R)-20
(R,R)-2
8.38 0.07
6.05 0.30
8.12 0.09
5.66 0.02
6.92 0.03
7.42 0.03
7.72 0.05
7.83 0.15
7.44 0.06
6.08 0.18
6.30 0.07
5.69 0.19
<5
5.83 0.04
6.11 0.08
5.92 0.01
5.92 0.10
6.16 0.06
199
0.5
250
—
12.5
20
63
80
20
selectivity towards D2 receptors ([35S]GTP
cS, pEC50 = 6.57 0.2,
Emax = 54%), further testing in animal models of Parkinson and
depression are currently under progress and results will be re-
ported in due course.
Ropinirole
Acknowledgments
a,b
See Table 1.
Single crystal X-ray diffraction was performed by Mrs. P. Retail-
leau at the Institut de Chimie des Substances Naturelles at Gif sur
Yvette (France). We are indebted to Mr. C. Nisole and Mrs. C. Scha-
effer for their help with spectroscopic and crystallographic data,
and to Ms. C. Rivalland for excellent secretarial assistance.
Table 3
Activation of hD3 receptors by 2, 15, 16, 19 as determined by an antibody capture/
scintillation proximity assay of [35S]GTP
S binding to Gai3
c
(R,R)-2
(R,R)-15
(R,R)-16
(R,R)-19
a
pEC50
Emax
8.64 0.07
8.85 0.04
74.2% 0.32
8.64 0.02
74% 5.00
8.33 0.17
34% 2.70
References and notes
68%
2
1. Vallone, D.; Picetti, R.; Borelli, E. Neurosci. Biobehav. Rev. 2000, 24, 125.
2. Sokoloff, P.; Giros, B.; Martres, M. P.; Bouthenet, M. L.; Schwartz, J. C. Nature
1990, 347, 146.
a
pEC50, ꢀlog(effective concentration50).
3. Joyce, J. N. Pharmacol. Ther. 2001, 90, 231.
4. Neve, K. A.; Seamans, J. K.; Trantham-Davidson, H. J. Recept. Signal Transduct.
Res. 2004, 24, 165.
5. Sokoloff, P.; Diaz, J.; Le Foll, B.; Guillin, O.; Leriche, L.; Bezard, E.; Gross, C. CNS
Neurol. Disord. Drug Targets 2006, 5, 25.
6. Millan, M. J.; Brocco, M. Therapie 2008, 63, 187.
7. Xi, Z. X.; Gardner, E. L. CNS Drug Rev. 2007, 13, 240.
8. Newman-Tancredi, A.; Cussac, D.; Audinot, V.; Nicolas, J. P.; De Ceuninck, F.;
Boutin, J. A.; Millan, M. J. J. Pharmacol. Exp. Ther. 2002, 303, 805.
9. Gallagher, D. A.; Schrag, A. CNS Drugs 2008, 22, 563.
10. Yamamoto, M.; Schapira, A. H. Expert. Rev. Neurother. 2008, 8, 671.
11. Whone, A. L.; Watts, R. L.; Stoessl, A. J.; Davis, M.; Reske, S.; Nahmias, C.; Lang,
A. E.; Rascol, O.; Ribeiro, M. J.; Remy, P.; Poewe, W. H.; Hauser, R. A.; Brooks, D.
J. REAL-PET Study Group. Ann. Neurol. 2003, 54, 93.
12. Chen, S.; Zhang, X.; Yang, D.; Du, Y.; Li, L.; Li, X.; Ming, M.; Le, W. FEBS Lett.
2008, 582, 603.
13. Millan, M. J.; Audinot, V.; Rivet, J. M.; Gobert, A.; Vian, J.; Prost, J. F.; Spedding,
M.; Peglion, J. L. Eur. J. Pharmacol. 1994, 260, R3.
14. Millan, M. J.; Peglion, J. L.; Vian, J.; Rivet, J. M.; Brocco, M.; Gobert, A.; Newman-
Tancredi, A.; Dacquet, C.; Bervoets, K.; Girardon, S. J. Pharmacol. Exp. Ther. 1995,
275, 885.
15. Gobert, A.; Rivet, J. M.; Audinot, V.; Cistarelli, L.; Spedding, M.; Vian, J.; Peglion,
J. L.; Millan, M. J. J. Pharmacol. Exp. Ther. 1995, 275, 899.
16. Peglion, J. L.; Vian, J.; Goument, B.; Despaux, N.; Audinot, V.; Millan, M. J. Bioorg.
Med. Chem. Lett. 1997, 7, 881.
17. Audinot, V.; Newman-Tancredi, A.; Gobert, A.; Rivet, J. M.; Brocco, M.; Lejeune,
F.; Gluck, L.; Despote, I.; Bervoets, K.; Dekeyne, A.; Millan, M. J. J. Pharmacol.
Exp. Ther. 1998, 287, 187.
18. Millan, M. J.; Cussac, D.; Gobert, A.; Lejeune, F.; Rivet, J. M.; Mannoury La Cour,
C.; Newman-Tancredi, A.; Peglion, J. L. J. Pharmacol. Exp. Ther. 2004, 309, 903.
creased affinity at hD3 sites and, accordingly, selectivity (compare
20 and 19). Finally, as for compounds 10–14, substitution on the
nitrogen atom of the lactam ring was deleterious for activity since
17, the N-methyl derivative of (R,R)-15, is devoid of activity.
Subsequently, it was necessary to verify the efficacy of the most
potent and selective derivatives (R,R)-15, 16 and 19. Indeed, the
adjunction of the extra ring may in theory have transformed ago-
nist into antagonist activity. For this purpose, we assessed the abil-
ity of these three compounds to stimulate hD3 receptors. Table 3
shows their potency and maximal efficacies for activation of hD3
receptors expressed relative to those of dopamine-induced facilita-
tion of [35S]GTP
cS binding to Gai3, a G-protein isoform preferen-
tially recruited by hD3 and hD2 receptors.19,35
The five-membered ring lactam (R,R)-15 and lactone (R,R)-16
are comparable in potency and maximal efficacy to the agonist
(R,R)-2, but the six-membered ring lactam (R,R)-19, albeit as po-
tent as (R,R)-2, is much less efficacious suggesting that the size
of the extra ring plays an important role in determining ligand
efficacy.
Replacement of the amide function of S32504 by an ester,
acid, tetrazole or cyano function improved neither affinity at
D3 receptors nor selectivity versus D2 receptors, while mono or

2138
J.-L. Peglion et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2133–2138
19. Millan, M. J.; Di Cara, B.; Hill, M.; Jackson, M.; Joyce, J. N.; Brotchie, J.; McGuire,
S.; Crossman, A.; Smith, L.; Jenner, P.; Gobert, A.; Peglion, J. L.; Brocco, M. J.
Pharmacol. Exp. Ther. 2004, 309, 921.
20. Millan, M. J.; Brocco, M.; Papp, M.; Serres, F.; Drieu La Rochelle, C.; Sharp, T.;
Peglion, J. L.; Dekeyne, A. J. Pharmacol. Exp. Ther. 2004, 309, 936.
21. Triflate 3 was prepared from compound 25 of Ref. 16. 25 was first reduced by
LiAlH4 and then transformed in triflate by using triflic anhydride.
22. Selnick, H. G.; Smith, G. R.; Tebben, A. J. Synth. Commun. 1995, 25, 3255.
23. Anelli, P. L.; Brocchetta, M.; Palano, D.; Visigalli, M. Tetrahedron Lett. 1997, 38,
2367.
31. Characterization of (R,R)-15ꢂHCl: mp 287–291 °C, ½a D20
ꢁ
+52.4 (c 1, MeOH). 1H
NMR (300 MHz, CDCl3) (d ppm): 0.92 (t, 3H), 1.53 (m, 2H), 1.64 (m, 1H), 2.29
(m, 3H), 2.46 (td, 1H), 2.82 (m, 1H), 2.89 (d, 1H), 3.00 (m, 2H), 3.95 (td, 1H),
4.09 (m, 1H), 4.35 (d, 1H), 4.38 (m, 2H), 6.68 (m, 1H), 7.16 (s, 1H), 8.09 (s, 1H).
32. Characterization of (R,R)-16ꢂHCl: mp 268–271 °C. 1H NMR (300 MHz, DMSO-
d6) (d ppm): 0.95 (t, 3H), 1.75 (sext, 2H), 2.10 (m, 1H), 2.55 (m, 1H), 3.15–2.95
(m, 3H), 3.45–3.15 (m, 3H), 3.60 (m, 1H), 4.25 (m, 2H), 5.10 (d, 1H), 5.40 (s, 2H),
7.50 (s, 1H), 7.80 (s, 1H), 11.90 (m, 1H).
33. Characterization of (R,R)-19ꢂHCl: mp 263–265 °C. 1H NMR (300 MHz, DMSO-
d6) (d ppm): 1.00 (t, 3H), 1.75 (m, 2H), 2.00 (m, 1H), 2.50 (m, 2H), 2.85 (m, 2H),
3.00 (m, 3H), 3.4–3.15 (m, 5H), 3.60 (m, 1H), 4.25 (m, 2H), 5.00 (d, 1H), 7.10 (s,
1H), 7.75 (s, 1H), 7.90 (s, 1H).
24. Ono, M.; Araya, I.; Todoroki, R.; Tamura, S. Chem. Pharm. Bull. 1990, 38,
1824.
25. Cabri, W.; Candiani, I.; Bedeschi, A.; Penco, S. J. Org. Chem. 1992, 57, 1481.
26. Duncia, J. V.; Pierce, M. E.; Santella, J. B., III J. Org. Chem. 1991, 56, 2395.
27. Peglion, J. L.; Harmange, J. C.; Millan, M. J.; Lejeune, F. E.P. 899267, 1999.
34. Fisher, L. E.; Caroon, J. M.; Russell Stabler, J. S.; Lundberg, S.; Muchowski, J. M. J.
Org. Chem. 1993, 58, 3643.
35. The activation of G
a
i3 subunit by hD3 receptors in CHO cells was quantified
28. Characterization of (R,R)-2: mp 198–199 °C, ½a D20
ꢁ
+65.8 (c 1, CHCl3). 1H NMR
using the classical [35S]GTP
c
S binding assay coupled to an antibody capture/
(300 MHz, DMSO-d6) (d ppm): 0.90 (t, 3H), 1.60–1.35 (m, 3H), 2.40–2.00 (m,
4H), 2.98–2.70 (m, 4H), 3.78 (m, 1H), 4.00 (dd, 1H), 4.20 (d, 1H), 7.14 (d, 1H),
7.15 (s, 1H), 7.66 (dd, 1H), 7.90 (s, 1H), 7.94 (s, 1H).
scintillation proximity assay (SPA) in 96-well plates (PerkinElmer Life
Sciences). Membranes from CHO cells expressing hD3 receptors were
incubated with increasing concentration of each compound for 30 min before
29. Crystallographic data for the structure in this Letter have been deposited in the
Cambridge Crystallographic Data Centre for small molecules as supplementary
publication number CCDC 713653. Copies of the data can be obtained, free of
charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.
30. Sato, N. Tetrahedron Lett. 2002, 43, 6403.
addition of a mouse monoclonal antibody specifically recognizing the G
subunits recruited by D3 receptors. After 1 h of incubation, the activation of
i3 proteins was detected by adding scintillation beads, coated with anti-
mouse secondary antibodies, which emitted light when stimulated by the
35S]GTP
S bound.
ai3
G
a
[
c