Design and synthesis of APTCs (aminopyrrolidinetricarboxylic acids): Identification of a new group III metabotropic glutamate receptor selective agonist

Article abstract of DOI:10.1016/j.bmcl.2006.06.062

A new family of mGlu receptor orthosteric ligands called APTCs was designed and synthesized using a parallel chemistry approach. Amongst 65 molecules tested on mGlu4, mGlu6 and mGlu8 subtypes, (2S,4S)-4-amino-1-[(E)-3-carboxyacryloyl]pyrrolidine-2,4-dicar

Full text of DOI:10.1016/j.bmcl.2006.06.062

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4856–4860
Design and synthesis of APTCs (aminopyrrolidinetricarboxylic
acids): Identification of a new group III metabotropic
glutamate receptor selective agonist
a,
a
a
Stephan Schann,a, Christel Menet, Paul Arvault, Geraldine Mercier,
*
´
a
a
a
b,c
´
Melanie Frauli, Stanislas Mayer, Nadia Hubert, Nicolas Triballeau,
Hugues-Olivier Bertrand,c Francine Acherb and Pascal Neuvillea
aFaust Pharmaceuticals, BIOPARC, Boulevard Sebastien Brant, 67400 Illkirch, France
bLaboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, UMR 8601-CNRS, Universite´ Rene´ Descartes-Paris V,
`
45 rue des Saints-Peres, 75270 Paris Cedex 06, France
cAccelrys SARL, 91893 Orsay Cedex, France
Received 12 May 2006; revised 16 June 2006; accepted 17 June 2006
Available online 7 July 2006
Abstract—A new family of mGlu receptor orthosteric ligands called APTCs was designed and synthesized using a parallel chemistry
approach. Amongst 65 molecules tested on mGlu4, mGlu6 and mGlu8 subtypes, (2S,4S)-4-amino-1-[(E)-3-carboxyacryloyl]pyrrol-
idine-2,4-dicarboxylic acid (8a06—FP0429) has been shown to be a full mGlu4 agonist and a partial mGlu8 agonist. In addition,
8a06 was shown to be selective versus group I and II mGlu subtypes. A possible explanation for this efficacy difference is proposed
by docking experiment performed with molecular model of the receptor.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Glutamate, the most commonly occurring neurotrans-
mitter in the CNS, plays a critical role in a large number
of physiological processes. It operates through two fam-
ilies of receptors: the ionotropic (iGlu) and the metabo-
tropic (mGlu) receptors. mGlu receptors have been
divided in three groups according to their sequence
homologies, pharmacological properties and signal
transduction pathways.1 Group III mGlu receptors in-
clude mGlu4, mGlu6, mGlu7 and mGlu8 receptors,
which are negatively coupled to cAMP formation. They
are mainly localised in the central nervous system with
the exception of mGlu6 which is found mostly in the ret-
ina. Their potential roles for the treatments of anxiety,
Parkinson’s disease and epilepsy have been suggested.2
competing with glutamate in the venus fly trap (VFT)
domain and allosteric modulators interacting with the
transmembrane domain.3 To date, only few orthosteric
agonists have been described for group III mGlu
receptors. The most representative ones are L-AP4, R,
S-PPG,4 S-DCPG5 and ACPT-I6 (Fig. 1). They have
low-micromolar to sub-micromolar potencies for mGlu4,
mGlu6 and mGlu8 and high-micromolar potency for
mGlu7 receptor.
Because of their simple amino acid structures, mGlu
orthosteric ligands are not subject to parallel chemistry.
To our knowledge, there is no report of chemical
approach offering rapid access to a large number of such
mGlu receptor ligands can be classified in two families
according to their site of interaction: orthosteric ligands
H
HOOC
NH2
HOOC
NH2
H
H2N
COOH
COOH
HOOC
NH2
Keywords: Metabotropic glutamate receptor; mGlu4; Excitatory
amino acids.
COOH
COOH
HOOC
PO3H2
PO3H2
*
Corresponding author. Tel.: +33 390 406 150; fax: +33 390 406
155; e-mail: sschann@faustpharma.com
Present address: Galapagos NV, Generaal de Wittelaan L11A3, 2800
Mechelen, Belgium.
Figure 1. Structures of standard group III mGlu receptor agonists.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.06.062

S. Schann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4856–4860
4857
polar derivatives. In this communication, we report the
design and the synthesis of a new orthosteric ligand fam-
ily named APTC (for aminopyrrolidine-tricarboxylic
acid), and the pharmacological characterization of
8a06 (FP0429), a hit showing micromolar potencies on
group III subtypes.
chromatography to afford the four isomers 4a–d. These
aminoesters were boc-protected (5a–d) and hydrogenol-
ysed to the corresponding pyrrolidines 6a–d. The four
protected platforms were capped in parallel with electro-
philic species to give 7(a–d)XX using solid supported
reagents and scavengers. Final deprotections yielded
the desired APTCs 8(a–d)XX.8
APTCs design is based on structural modifications of
tricarboxylic acid ACPTs, family of molecules in which
selective group III agonists were found.6 The aim was
to obtain a closed scaffold, that is, an a-amino acid
on a five-membered ring with highly polar side chains
attached to positions 3 and 4. These side chains could
interact with the highly basic distal binding pocket of
the receptors. In addition, our objective was also to
obtain a new scaffold amenable to parallel synthesis.
Just like ACPTs are ACPD’s analogues with an
additional COOH group, we anticipated that function-
alizing APDCs on the pyrrolidine nitrogen with polar
groups would afford the expected family of ligands
(Fig. 2).
Using 25 different R groups (Scheme 1), chosen in order
to introduce an additional polar group on the N-1 posi-
tion, a virtual library of one hundred APTCs was
designed. Half of these groups is made of a carboxylic
acid, where as the other half contains non-acidic polar
functions. Synthetic efforts led to the synthesis of 65
final compounds giving a success rate of 65% for the
library synthesis.
Synthesized APTCs were screened on three group III
mGlu receptors (mGlu4, mGlu6 and mGlu8) in func-
tional assays at 450 and 300 lM, for their agonist and
antagonist activities, respectively.9 8a06 (Fig. 3) was
In APTCs, two asymmetric centres are present. We
decided to control both by choosing the stereochemical
synthesis depicted in Scheme 1. Starting material was
either 2S,4R or 2R,4R hydroxyproline which were
converted into hydantoins 3a and 3b following similar
procedures used by Monn et al. for the synthesis of
APDCs.7 Diastereoisomers 3a and 3b were then
hydrolysed, esterified and separated using flash
COOH
H2N
FP0429 (8a06)
N
COOH
HOOC
O
Figure 3. 8a06 structure.
H2N
COOH
COOH
H2N
COOH
COOH
H2N
COOH
COOH
H2N
COOH
COOH
N
N
H
HOOC
R
Figure 2. Design of APTC family.
O
H
N
OH
O
COOMe
H2N
O
HN
a-c
d
e
f
R groups:
N
N
N
H
Ph
N
COOMe
Ph
COOH
COOEt
13 - COPh
14 - COCH2OH
01 - COCOOH
02 - COCH2COOH
03 - CO(CH2)2COOH
04 - CO(CH2)3COOH
05 - CO(CH2)4COOH
Ph
COOEt
4a: 2S,4S
4b: 2S,4R
4c: 2R,4S
4d: 2R,4R
15 - COPh(m-OH)
16 - COPh(p-OMe)
17 - CO(N-morpholinyl)
18 - CO(3-pyridinyl)
19 - COOMe
1a: 2S
1b: 2R
2a: 2S
2b: 2R
3a: 2S
3b: 2R
06 - COCH=CHCOOH (E)
07 - COCH=CHCOOH (Z)
08 - COPh(o-COOH)
09 - COPh(m-COOH)
10 - COPh(p-COOH)
11 - CH2COOH
boc
HN
boc
HN
boc
HN
COOH
COOH
20 - COMe
H2N
COOMe
COOMe
COOMe
COOMe
COOMe
21 - CONHCOOEt
22 - CO(2-furanyl)
23 - CO(2-pyrroyl)
24 - SO2Et
g
h
i-j
N
R
N
N
H
N
R
Ph
COOMe
12 - CH2CH2COOH
25 - SO2NMe2
8aXX: 2S,4S
8bXX: 2S,4R
8cXX: 2R,4S
8dXX: 2R,4R
5a: 2S,4S
6a: 2S,4S
7aXX: 2S,4S
5b: 2S,4R
5c: 2R,4S
5d: 2R,4R
6b: 2S,4R
6c: 2R,4S
6d: 2R,4R
7bXX: 2S,4R
7cXX: 2R,4S
7dXX: 2R,4R
XX refers to R groups as defined in table above
Scheme 1. Synthesis of APTCs. Reagents and conditions: (a) SOCl2, EtOH, reflux, 95–99%; (b) PhCH2Br, Et3N, DCM, reflux, 79–90%; (c)
1—Oxalyl chloride, DCM, DMSO, À78 ꢁC, 2—Et3N, À78 ꢁC–rt, 92–96%; (d) (NH4)2CO3, KCN, EtOH, H2O, 50–55 ꢁC, 68–78%; (e) 1—NaOH
2 M, reflux, 2—SOCl2, MeOH, reflux, 3—diastereoisomer separation by standard chromatography, 7–15% for cis diastereoisomers, 30–60% for trans
diastereoisomers; (f) Boc2O, DCM, rt, 67–95%; (g) NH4HCO2, Pd-C 10%, MeOH, reflux, 57–95%; (h) 1—electrophilic agent, PS-DEA, DCM, rt,
2—AMPS, DCM, rt; (i) LiOH 2 M, THF, rt; (j) HCl 2 M in ether, AcOH, rt, 5–50% for the three last steps.

4858
S. Schann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4856–4860
detected with strong agonist responses on mGlu4 and
lower one on mGlu6 and mGlu8 at 450 lM.10 It is inter-
esting to notice that amongst inactive APTCs on
mGlu4, some are structurally very close to 8a06. It is
the case of 8a03 or 8d06 which differ only by saturation
of the double bond or stereochemistry of the pyrrolidine
position 2, respectively. This result further illustrates the
particularly restricting feature of group III mGlu active
sites.
In order to further understand the interaction of 8a06
with mGlu4, a docking experiment was performed using
the receptor molecular model and docking protocol
recently described by Bertrand et al.11 This protocol
consists in the flexible fitting of the ligand within a rigid
receptor using the shape-based docking algorithm
LigandFit. The obtained poses were subsequently scored
using the LigScore scoring function. The best pose was
then energy minimized with CHARMm allowing full
flexibility for the ligand and only side-chain flexibility
for the receptor. All calculations were carried out in
the Discovery Studio 1.5 environment (Accelrys, San
Diego, California). Results show that 8a06 fits nicely
in a dense H-bond network, interacting with a dozen
amino acids of mGlu4 active site. The glycine entity of
8a06 is bound to the set of residues that build up the sig-
nature motif common to all proteins of the fold family
(S159, A180, T182, D202, Y230, D312).11,12 The distal
functions of 8a06, one amide and two acidic groups,
are bound to the same residues as ACPT-I one’s (K71,
K74, R78, S110, S157, R258, N286, S313, K317,
K405)11 (Fig. 5). Yet, because of conjugation, the
fumaric acid amide substituent is rigid, thus when all po-
lar bindings are fitted, one of the ethylenic protons is
We then determined EC50 values for mGlu4, 6 and 8 and
selectivity against other mGlu subtypes (Table 1). 8a06
activates both mGlu4 and mGlu8 with micromolar
potency (EC50 = 48.3 lM and EC50 = 56.2 lM, respec-
tively) and mGlu6 with high micromolar potency
(EC50 = 378 lM). It shows also selectivity for group
III mGlu receptors as no agonist nor antagonist activity
was detected on subtypes mGlu1, mGlu5 and mGlu2 up
to 5 mM. Interestingly, although 8a06 shares a similar
potency on mGlu4 and mGlu8, its efficacy for the two
subtypes differs significantly. 8a06 is a full agonist on
mGlu4 with a maximal response almost comparable to
that of glutamate (Emax = 75%) and a partial agonist
on mGlu8 with a Emax value of 36% of glutamate
response (see Fig. 4).
˚
found in close contact with those of G158 (2.6 A)
Table 1. Activities of 8a06 and standard group III agonists on rat mGlu receptors (EC50 values in lM)9
Group I Group II
Group III
mGlu7
NT
mGlu1
mGlu5
mGlu2
mGlu3
mGlu4
mGlu6
mGlu8
8a06a
>5000
>1000
>500
>5000
>1000
>500
>5000
>1000
>300
NT
>1000
>200
>100
NT
48.3 (5.2)
0.5–1
5.2
378 (112)
0.6–0.9
4.7
56.2 (14.6)
0.6
0.21
L-AP4b
RS-PPGc
S-DCPGd
ACPT-I
160–800
185
ant @ 32
ant @ 1000e
>100
>1000a
>100
>1000e
8.8
7.2e
3.6
18.4f
>100
1200f
0.031
10.1f
NT: not tested.
a Values are means of three independent experiments, standard deviation is given in parentheses.
b Taken from Refs. 3 and 13.
c On human mGlu, taken from Ref. 4.
d On human mGlu, taken from Ref. 5.
e Taken from Ref. 6.
f Personal communication from Dr. Jean-Philippe Pin.
Basal
30000
25000
10000
Glu max
8a06
8000
20000
6000
15000
4000
10000
2000
5000
0
0
rat mGlu4
rat mGlu8
Figure 4. 8a06 efficacies on mGlu4 and mGlu8. Intracellular calcium mobilization in HEK 293 expressing the indicated receptor, under basal
conditions (white bars), 500 lM glutamate (grey bars) and 1 mM 8a06 (black bars) activations. No activity was detected with either Glutamate or
8a06 on mock-transfected cells.

S. Schann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4856–4860
4859
Sci. U.S.A. 2003, 100, 13668; (d) Valenti, O.; Marino, M.
J.; Wittmann, M.; Lis, E.; DiLella, A. G.; Kinney, C. G.;
Conn, P. J. J. Neurosci. 2003, 23, 7218; (e) Moldrich, R.
X.; Chapman, A. G.; De Sarro, G.; Meldrum, B. S. Eur. J.
Pharmacol. 2003, 476, 3.
3. (a) Schoepp, D. D.; Jane, D. E.; Monn, J. A. Neurophar-
macology 1999, 38, 1431; (b) Pin, J.-P.; De Colle, C.;
Bessis, A.-S.; Acher, F. C. Eur. J. Pharmacol. 1999, 375,
277; (c) Kew, J. N. Pharmacol. Ther. 2004, 104, 233; (d)
Sabbatini, F. M.; Micheli, F. Exp. Opin. Ther. Patents
2004, 14, 1593; (e) Shipe, W. D.; Wolkenberg, S. E.;
Williams, D. L.; Lindsley, C. W. Curr. Opin. Drug Disco.
Dev. 2005, 449.
4. Gasparini, F.; Bruno, V.; Battaglia, G.; Lukic, S.; Leon-
hardt, T.; Inderbitzin, W.; Laurie, D.; Sommer, B.;
Varney, M. A.; Hess, S. D.; Johnson, E. C.; Kuhn, R.;
Urwyler, S.; Sauer, D.; Portet, C.; Schmutz, M.; Nicoletti,
F.; Flor, P. J. J. Pharmacol. Exp. Ther. 1999, 290, 1678.
5. Thomas, N. K.; Wright, R. A.; Howson, P. A.; Kinston,
A. E.; Schoepp, D. D.; Jane, D. E. Neuropharmacology
2001, 40, 311.
6. Acher, F. C.; Tellier, F. J.; Azerad, R.; Brabet, I. N.;
Fagni, L.; Pin, J.-P. J. Med. Chem. 1997, 40, 3119.
7. Monn, J. A.; Valli, M. J.; Johnson, B. G.; Salhoff, C. R.;
Wright, R. A.; Howe, T.; Bond, A.; Lodge, D.; Spangle, L.
A.; Paschal, J. W.; Campbell, J. B.; Griffey, K.; Tizzano, J.
P.; Schoepp, D. D. J. Med. Chem. 1996, 39, 2990.
8. Typical experimental procedure for capping and depro-
tections of intermediates 6a–d: Electrophilic species
(0.33 mmol) was added to a solution of 6a–d (50 mg,
0.165 mmol) and PS-DEA (200 mg, 0.33 mmol) in DCM
(2 mL). The resulting suspension was shaken for 20 h.
Then AMPS (200 mg, 0.33 mmol) was added to the
solution, and this latter was shaken for a further 20 h.
Resins were filtered and the solvent was evaporated.
Lithium hydroxide (2 M solution, 0.3 mL) was added to a
solution of crude ester 7(a–d)XX in THF (1 mL). The
resulting mixture was stirred at RT for 12 h, then acidified
to pH 1 with 1 M hydrochloric acid, extracted with
AcOEt, washed with brine, dried over MgSO4 and
evaporated. Hydrochloric acid (2 M solution in ether,
0.2 mL) was added to a solution of crude boc-amine
8(a–d)XX in acetic acid (0.5 mL). The resulting solution
was stirred for 6 h at RT. Solvents were evaporated and
the resulting solid was suspended in ether, filtered and
dried to give expected product as a white solid. All
Figure 5. Agonist 8a06 docked at mGlu4 active site.
(Fig. 5). Consequently, any bulkier residue as an alanine
would induce some distortion of the binding network.
Indeed, the corresponding amino acid of G158 in mGlu6
and mGlu8 is no longer a glycine but an alanine. In
addition, S157 to which 8a06 is also bound (Fig. 5) is
replaced by an alanine in mGlu8. These differences
may be responsible for the different behaviour observed
with 8a06 on these subtypes. The latter assumption
should be validated by using mGlu4, 6 and 8 mutants
where the crucial amino acids would be replaced.
In conclusion, a chemical library of new ligands was
designed and synthesized using parallel chemistry
approach. An agonist for group III was identified with
different efficacies for subtypes mGlu4 and mGlu8. This
ligand will be of great interest to further understand
pharmacological role and therapeutic potential of
mGlu4 subtype.
1
compounds were identified by H NMR and LC/ES–MS
Acknowledgments
and data were consistent with the proposed structures.
9. Intracellular calcium measurements: HEK 293 cells were
transiently transfected with plasmid DNA encoding rat
mGlu receptors, and a chimeric G protein to couple the
naturally Gi/o coupled receptors (subtypes 2, 4, 6 and 8) to
the Phospholipase C/Ca2+ pathway (Frauli, M.; Neuville,
P.; Vol, C.; Pin, J.-P.; Prezeau, L. Neuropharmacology
2006, 50, 245). In summary, cells were seeded in poly-
ornithine-coated, black-walled, clear bottomed, 96-well
plates and cultured for 24 h after transfection. Cells were
then washed with freshly prepared buffer (1· HBSS
supplemented with 20 mM Hepes, 1 mM MgSO4,
3.3 mM Na2CO3, 1.3 mM CaCl2, 0.5% BSA and 2.5 mM
Probenecid) and loaded with 1 lM Ca2+-sensitive fluores-
cent dye Fluo4AM (Molecular Probes). After loading,
cells were washed twice with buffer and then incubated in
50 lL of buffer. Addition of compounds (50 ll of 2·-drug
solution) and intracellular Ca2+ measurements were per-
formed by the fluorescence microplate reader FlexStation
(Molecular Devices) at sampling intervals of 1.5 s for 60 s
(excitation 485 nm, emission 525 nm). Agonist or antag-
onist activities were evaluated in comparison to basal
The authors thank Dr. Jean-Philippe Pin and Dr. Anne-
Sophie Bessis for completing data on ACPT-I pharma-
´
cological characterization. We also thank Dr. Andre
Mann, Angele Schoenfelder and Dr. Philippe Klotz for
`
their assistance in the chemical route design.
References and notes
1. (a) Conn, P. J.; Pin, J.-P. Annu. Rev. Pharmacol. Toxicol.
1997, 37, 205; (b) Pin, J.-P.; Acher, F. C. Curr. Drug
Targets CNS Neurol. Disord. 2002, 1, 297.
2. (a) Stachowicz, K.; Klak, K.; Klodzinska, A.; Chojna-
Wojcik, E.; Pilc, A. Eur. J. Pharmacol. 2004, 498, 153; (b)
Palucha, A.; Tatarczynska, E.; Branski, P.; Szewczyk, B.;
Wieronska, J. M.; Klal, K.; Chojna-Wojcik, E.; Pilc, A.
Neuropharmacology 2004, 46, 151; (c) Marino, M. J.;
Williams, D. L., Jr.; O’Brien, J. A.; Valenti, O.; McDon-
ald, T. P.; Clements, M. K.; Wang, R.; DiLella, A. G.;
Hess, J. F.; Kinney, C. G.; Conn, P. J. Proc. Natl. Acad.

4860
S. Schann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4856–4860
signal or signal evoked by Glutamate EC80 alone,
d, J = 14.0 Hz), 3.02 (1Hminor, m), 2.93 (1Hmajor, dd,
J = 9.9 and 14.6 Hz), 2.47 (1Hminor, m), 2.93 (1Hmajor, dd,
J = 4.4 and 14.0 Hz). ES–MS (m/z): 273 (M+H)+.
11. Bertrand, H. O.; Bessis, A.-S.; Pin, J.-P.; Acher, F. C. J.
Med. Chem. 2002, 45, 3171.
respectively. Dose–response curves were fitted using the
GraphPad Prism program (San Diego). All values are
means SEM of three independent experiments per-
formed in triplicate.
10. Compound 8a06 was resynthesized using non-parallel
procedure. Characteristic data for 8a06: H NMR (D2O,
12. Acher, F. C.; Bertrand, H. O. Biopolymers 2005, 80, 357.
13. (a) Wu, S.; Wright, R. A.; Rockey, P. K.; Burgett, S. G.;
Arnold, P. R.; Johnson, B. G.; Schoepp, D. D.; Belagaje,
R. Mol. Brain Res. 1998, 53, 88; (b) Laurie, D. J.;
Schoeffter, P.; Wiederhold, K. H.; Sommer, B. Neuro-
pharmacology 1997, 36, 145.
1
400 MHz) d mixture of rotamers, 7.21 (1Hmajor, d,
J = 15.5 Hz), 7.10 (1Hminor, d, J = 15.5 Hz), 6.68 (1Hmajor
,
d, J = 15.2 Hz), 6.65 (1Hminor, d, J = 15.5 Hz), 4.58 (1H,
m), 4.27 (1H, d, J = 12.0 Hz), 4.13 (1H, m), 3.94 (1Hminor
,