Synthesis of N1-substituted analogues of (2R,4R)-4-amino-pyrrolidine-2,4-dicarboxylic acid as agonists, partial agonists, and antagonists of group II metabotropic glutamate receptors

Article abstract of DOI:10.1016/S0960-894X(01)00329-8

The chemical synthesis of a series of N¹-substituted derivatives of (2R,4R)-4-aminopyrrolidine-2,4-dicarboxylic acid [(2R,4R)-APDC] as constrained analogues of γ-substituted glutamic acids is described. Appropriate substitution of the N¹-position results in agonist, partial agonist, or antagonist activity at mGluR2, mGluR3, and/or mGluR6.

Full text of DOI:10.1016/S0960-894X(01)00329-8

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1919–1924
Synthesis of N1-Substituted Analogues of (2R,4R)-4-Amino-
pyrrolidine-2,4-dicarboxylic Acid as Agonists, Partial Agonists,
and Antagonists of Group II Metabotropic Glutamate Receptors
Jayanta Kumar Mukhopadhyaya,a Alan P. Kozikowski,a,* Ewa Grajkowska,b
Sergey Pshenichkinb and Jarda T. Wroblewskib
aDepartment of Neurology, Drug Discovery Program, Georgetown University Medical Center, 3900 Reservoir Road,
N.W., Washington, DC 20007, USA
bDepartment of Pharmacology, Georgetown University Medical Center, 3900 Reservoir Road,
N.W., Washington, DC 20007, USA
Received 2 March 2001; accepted 7 May 2001
Abstract—The chemical synthesis of a series of N1-substituted derivatives of (2R,4R)-4-aminopyrrolidine-2,4-dicarboxylic acid
[(2R,4R)-APDC] as constrained analogues of g-substituted glutamic acids is described. Appropriate substitution of the N1-position
results in agonist, partial agonist, or antagonist activity at mGluR2, mGluR3, and/or mGluR6. # 2001 Elsevier Science Ltd. All
rights reserved.
The amino acid glutamate plays a pivotal role in bio-
logical processes ranging from memory and learning to
neuronal degeneration. This major excitatory amino
acid (EAA) acts through disparate glutamate receptors,
which can be categorized into two distinct types, the so-
called ionotropic receptors (iGluRs) and the metabo-
tropic receptors (mGluRs).1 iGluRs are associated with
integral cation-specific ion channels and include the N-
methyl-d-aspartate (NMDA), 2-amino-3-(5-methyl-3-
hydroxyisoxazol-4-yl)propanoic acid (AMPA), and kai-
nate subtypes. On the other hand, mGluRs are coupled
to cellular effectors through GTP-binding proteins.
Pharmacologically, mGluRs have been distinguished
from the iGluRs by the use of the mGluR-selective
agonist, (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic
acid [(1S,3R)-ACPD], generally through measurements
PI hydrolysis, whereas group II (mGluR2 and 3) and
group III (mGluR4, 6, 7, and 8) are negatively linked to
adenylyl cyclase activity. The group I receptors are
more sensitive to quisqualic acid than they are to
ACPD, the group II receptors are more sensitive to
ACPD than to quisqualic acid, and the group III
receptors are most sensitive to 2-amino-4-phosphono-
butyric acid (l-AP4).2
In order to better characterize the roles of GluRs in
physiological processes,3À5 there is an important need to
identify novel, high affinity ligands that are family and
subtype specific.6 While a number of mGluR selective
compounds (Scheme 1) have been described to date,
such as LY354740 and its heteroatom analogues,7
APDC,8 ABHxD-I,9 1-benzyl-APDC,10 and 1-amino-
APDC,11 the goal of having agonists, partial agonists,
and antagonists of exquisite selectivity for each of the
known subtypes has not been fully achieved. Herein we
report on additional N1-substituted analogues of (2R,4R)-
4-aminopyrrolidine-2,4-dicarboxylic acid (APDC). Of
interest is our findingthat dependingupon the nature of
the ringnitrogen substitutent, one is able to identify
ligands capable of acting as agonists, partial agonists, or
antagonists at mGluR2 while maintaining the same
rigidified glutamate template, namely APDC. As APDC
itself is a full agonist of the group II receptors,8 the
2+
involvingphosphoinositide (PI) hydrolysis or Ca
mobilization. To date the use of expression cloning
techniques has led to the identification of eight mGluR
subtypes, which have been placed into three major
categories based on their molecular structure, signal
transduction mechanisms, and pharmacological proper-
ties. Group I mGluRs (mGluR1 and 5) are coupled to
*Correspondingauthor. Tel.: +1-202-687-0686; fax: +1-202-687-
5065; e-mail: kozikowa@georgetown.edu
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00329-8

1920
J.K.Mukhopadhyaya et al./ Bioorg.Med.Chem.Lett.11 (2001) 1919–1924
present findingopens up another venue for mGluR
compound design.
hydrogenolysis over Pd(OH)2/C to give 3 which is
the key intermediate for the preparation of the various
N1-substituted APDC derivatives.
We have previously reported on synthetic routes to 1-
amino-APDC and to 1-benzyl-APDC, and much of the
same chemistry was employed in the preparation of the
compounds detailed herein. The synthesis of com-
pounds 7a–b, 8, and 9a–o is depicted in Scheme 2. The
Alkylation at the pyrrolidine ringnitrogen atom ( N1)
with a variety of alkylatingagents (Scheme 2) was car-
ried out under standard conditions and afforded inter-
mediates 4a–b, 5, and 6a–o in good yield. The alkylating
agent 2-(phenylethynyl)benzyl bromide required for the
synthesis of 6o was prepared by coupling2-bromo-
benzaldehyde with phenylacetylene under Pd(0) catalysis
intermediate 2 was prepared startingfrom
cis-4-
hydroxy-d-proline (1) followingthe previously reported
procedure.11,12 Compound 2 was debenzylated by
Scheme 1. Some typical ligands exhibiting mGluR activity.
Scheme 2. Synthesis of N1-substituted analogues of (2R,4R)-4-aminopyrrolidine-2,4-dicarboxylic acid. Reagents and conditions: (i) see refs 11 and
12; (ii) Pd(OH)2/C, H2 (1 bar), MeOH, 4 h; (iii) RCH(Br)COOMe, i-Pr2NEt, CH2Cl2, rt; (iv) RBr, i-Pr2NEt, CH2Cl2, rt (examples 6a–b, 6e, 6g–k,
and 6o) or RCl, i-Pr2NEt, Bu4NI (20 mol%), CH2Cl2, rt (examples 6c–d, 6f, and 6l); (v) 6 N HCl, reflux, 2–3 h; (vi) 1 N NaOH, THF, rt; (vii) 6 N
HCl, rt; (viii) NH2NH2, Raney-Ni, H2O, 35–40 ꢀC, 1 N HCl, for the conversion of 9i and 9k to 9m and 9n, respectively: (ix) 2-pyridylmethyl
mesylate, i-Pr2NEt, CH2Cl2, rt.

J.K.Mukhopadhyaya et al./ Bioorg.Med.Chem.Lett.11 (2001) 1919–1924
1921
followed by sequential reduction of the aldehyde func-
tionality with sodium borohydride and conversion of
the alcohol to bromide with phosphorus tribromide.
For the synthesis of compound 5, the required alkylat-
ingagent, 2-pyridylmethyl mesylate, was prepared from
2-pyridylcarbinol by reaction with methanesulfonyl
chloride in the presence of triethylamine. Other alkylat-
ingaegnts that are not commercially available were
prepared from the correspondingalcohols with phos-
phorus tribromide. The target amino acids 7b, 9a–b, and
9e–l were isolated as their hydrochloride salts in quan-
titative yield from 4b, 6a–b, and 6e–l by direct hydrolysis
with 6 N HCl under reflux conditions. Direct acidic
hydrolysis of compounds 4a, 5, 6c, 6d, and 6o afforded
inseparable mixtures of the desired compounds and the
correspondingdebenzylated products. To avoid this
problem, these intermediates were subjected to sequen-
tial hydrolysis of the ester and Boc protectinggroups
with 1 N NaOH in THF followed by treatment with
dilute HCl (pH 1) to produce 7a, 8, 9c, 9d, and 9o. The
nitro derivatives 9i and 9k were additionally converted
to the correspondingamines 9m and 9n usingthe Raney
nickel–hydrazine system followed by acidification with
dilute HCl. All final products and synthetic inter-
mediates13 exhibited satisfactory spectral and analytical
(C, H, N) data.
measurement by determination of PI hydrolysis.15
Within each experiment, the results were normalized to
the maximal response induced by 1 mM glutamate for
mGluR1 and mGluR5, 100 mM glutamate for mGluR2,
100 mM APDC for mGluR3, and by 100 mM 4-amino-
phosphonobutyrate (AP4) for mGluR4, mGluR6, and
mGluR8, and are expressed as percent of the maximal
response. Antagonist activity was measured in the
presence of the above agonists used at concentrations
yieldinghalf-maximal stimulation (EC 50). The evalu-
ation of the relative potencies of the tested compounds
and the determination of their EC50 or IC50 values was
performed by fittingthe normalized data to the logistic
equation by nonlinear regression.
Initially, all compounds were tested at 100 mM con-
centrations for agonist and antagonist activity at all
seven mGluRs. None of the compounds showed sig-
nificant activity (>20%) at mGluR1, mGluR5,
mGluR4, or mGluR8 receptors. In contrast, many of the
APDC derivatives were active as agonists or antagonists
at both group II receptors: mGluR2 (Fig. 1) and
mGluR3 (Fig. 2). Several compounds showed also
agonist activity at the mGluR6 receptor (Fig. 3). Results
from the dose–response experiments are summarized in
Table 1. As will be noted, agonist, antagonist, and par-
tial agonist activity is shown by some of these ligands, and
such activity is dependent upon subtle variations in the
nature of the functionality present in the N1-substituent.
The activity of these new APDC derivatives was tested
in seven cell lines with stable expression of the indivi-
dual subtypes of metabotropic glutamate receptors.
mGluR1a and mGluR5a (group I) receptors were
expressed in Chinese hamster ovary (CHO) cells,14 and
activity at these receptors was measured by the assay of
agonist-stimulated phosphoinositide (PI) hydrolysis.9
mGluR2 (group II) and mGluR6 (group III) receptors
were expressed in CHO cells, while the mGluR4 recep-
tor (group III) was expressed in baby hamster kidney
(BHK) cells.14 Activity at these receptors was deter-
mined by measurement of their ability to decrease the
forskolin-induced elevation of cAMP formation.9
Receptors mGluR3 (group II) and mGluR8 (group III)
were expressed in CHO cells in the presence of a chi-
meric G protein (Gqi9) which allows the couplingof
these receptors to phospholipase C and, hence, activity
We have previously reported on the pharmacology of 1-
amino-APDC11 and 1-benzyl-APDC.10 The former
ligand is a partial agonist of group II receptors. At both
receptors, the maximal stimulation reaches about 50%
of the stimulation obtained with full agonists. The EC50
values were found to be 5.5Æ2.1 mM for mGluR2 and
39Æ11 mM for mGluR3. On the other hand, the 1-ben-
zyl derivative of APDC showed selectivity very different
from that reported for APDC. Tested on CHO cells
expressingthe various mGluRs, 1-benzyl-APDC was an
agonist of mGluR6 with an EC50 of 20 mM. In contrast
to APDC, 1-benzyl-APDC was not an agonist of
mGluR2, in fact it behaved as a weak antagonist with
an IC50 of 200 mM. Additionally, and subsequent to our
Figure 1. Activity of APDC analogues acting as agonists (A) and antagonists (B) in cells expressing mGluR2 receptors. Activity of antagonists was
tested in the presence of 2 mM glutamate. Points represent means and error barsÆSEM from at least three experiments performed in triplicate.

1922
J.K.Mukhopadhyaya et al./ Bioorg.Med.Chem.Lett.11 (2001) 1919–1924
Figure 2. Activity of APDC analogues acting as agonists (A) and antagonists (B) in cells expressing mGluR3 receptors. Activity of antagonists was
tested in the presence of 2 mM APDC. Points represent means and error barsÆSEM from at least three experiments performed in triplicate.
As can be noted from Table 1, all of the newly synthe-
sized compounds are less potent than APDC at group II
mGluRs. Of the compounds presented, analogue 7a,
which contains a carboxymethyl group on the ring
nitrogen, is particularly interesting, as this compound
behaves as a partial agonist at mGluR2 with an appar-
ent EC50 value of 610 nM, while beinginactive at all
other mGluRs. Curiously, the only other compounds to
show this partial agonistic character at mGluR2 are the
4-carboxybenzyl derivative 9h and the 4-aminobenzyl
derivative 9n. Like 7a, 9h is also a triacid, but with a
larger spacer between the ring nitrogen and the CO2H
group. Its potency is lower than that of 7a, with an EC50
value of 4.7 mM. Compound 9h also shows weak agonist
activity at mGluR6 with an EC50 of 58 mM.
Figure 3. Activity of APDC analogues acting as agonists in cells
expressingmGluR6 receptors. Points represent means and error
barsÆSEM from at least three experiments performed in triplicate.
Of the various derivatives showingfull agonist activity,
the most potent of these at mGluR2 are 9a and 9i,
which bear a 2-hydroxybenzyl (EC50=6.1 mM) or 2-
nitrobenzyl group (EC50=4 mM), respectively. Com-
pound 9i also shows agonist activity at mGluR3. The 2-
aminobenzyl and 2-carboxybenzyl derivatives (9m and
report on 1-benzyl-APDC, researchers at the Lilly
Company published accounts of their own studies on
the activity of a number of monochlorobenzyl, dichloro-
benzyl, diphenylalkyl, naphthylmethyl, and phenylalkyl
derivatives of APDC, showingthat some of these
ligands were capable of acting as reasonably potent
group II mGluR antagonists.16 Interestingly, none of
the compounds in the Lilly series showed any agonist
activity. Thus, in the present study, we chose to extend
the SAR of these 1-substituted derivatives of APDC,
with the idea to further explore the dependence of their
biological activity on the nature of the N1-substituent.
We considered the possibility that an improved potency
as well as selectivity might be achieved through varia-
tions of the electronic character of the benzene ringby
the attachment of appropriate functionality. Addition-
ally, we considered the possibility that the aryl ring
substituent might influence agonist versus antagonist
activity, based upon the idea that the substituent may
either prevent or promote closure of the two lobes that
make up the amino terminal domain of the bindingsite
for the mGluRs.17
9f) also function as agonists, but they are much less
potent. However, in contrast to these results, it is inter-
estingto find that compounds 9d, 9e, 9j, and 9o all act
as antagonists or partial antagonists at mGluR2 and
mGluR3. Thus, while the 4-hydroxybenzyl derivative
9c, which contains an extra H-bond donor group,
behaves as a weak agonist, the corresponding methoxy
derivative is an antagonist. Of the two ortho-substituted
antagonists, 9o bearing the larger phenylethynyl group
is slightly more potent than the 2-bromo derivative 9e.
The pyridylmethyl analogue 8, which can be considered
to be a bioisostere of 9a, functions as an agonist at
mGluR2, albeit of lower potency. Lastly, compound 7b,
while actingas an antagonist of mGluR2 and mGluR3,
shows the best mGluR6 agonist activity.
In summary, as is apparent from the new analogues
reported herein, it is possible to alter the pharmacolog-
ical profile of APDC to arrive at structures capable of
functioning as agonists, partial agonists, or antagonists
at mGluR2. This alteration of activity is effected
through the choice of the substituent borne by the ring

J.K.Mukhopadhyaya et al./ Bioorg.Med.Chem.Lett.11 (2001) 1919–1924
1923
Table 1. Pharmacological characterization of N1-substituted analogues of APDC at mGluR2, mGluR3, and mGluR6
Compound
MGluR2 (mMa)
Agonists EC50 Antagonists IC50
mGluR3 (mMa)
Antagonists IC50
mGluR6 (mMa)
Agonists EC50
Agonists EC50
APDC
7a
7b
0.48
0.61b
1.7
na
na
na
22
na
21
51
8
9a
9b
9c
37
6.1
950
37
na
27
na
88b
na
na
7.5b
24
74
9d
9e
9f
9g
9h
9i
9j
9k
9l
9m
9n
9o
MCCGd
MCPGd
110
77
190
77
na
na
na
320
310
na
57
58
110
100
na
210
na
33
na
na
na
6.8
na
na
4.7b
4.0
45
na
93
na
na
na
na
23
160
na
35
2.3b
na
na
na
na
na
na
na
na
45c
62
68
270
290
aValues were calculated by nonlinear regression from data shown in Figures 1–3 (na=not active).
bPartial agonist.
cMaximal inhibition reached only 50%.
dTwo antagonists, a-methyl-2-(carboxycyclopropyl)glycine (MCCG) and a-methyl-4-carboxyphenylglycine (MCPG), were used for comparison.
9. Kozikowski, A. P.; Steensma, D.; Araldi, G. L.; Tuckman-
tel, W.; Wang, S.; Pshenichkin, S.; Surina, E.; Wroblewski,
J. T. J.Med.Chem. 1998, 41, 1641.
10. Tuckmantel, W.; Kozikowski, A. P.; Wang, S.; Pshe-
nichkin, S.; Surina, S.; Wroblewski, J. T. Bioorg.Med.Chem.
Lett. 1997, 7, 601.
11. Kozikowski, A. P.; Araldi, G. L.; Tuckmantel, W.; Pshe-
nichkin, S.; Surina, E.; Wroblewski, J. T. Bioorg.Med.Chem.
Lett. 1999, 9, 1721.
12. Tanaka, K.; Sawanishi, H. Tetrahedron: Asymmetry 1995,
6, 1641.
nitrogen of APDC. Based upon the recently published
X-ray structure of the extracellular ligand-binding
region of mGluR1,18 one may suggest that these sub-
stituents are capable of alteringthe dynamic equilibrium
amongthe various conformational states of the receptor,
so as to favor the open, closed (active), or intermediate
conformational states. Homology-based modeling
methods may allow for a more precise understandingof
these substituent effects.19
13. Representative procedure: To a solution of 3 (170 mg,
0.562 mmol) and 2-nitrobenzyl bromide (134 mg, 0.618 mmol)
in dichloromethane (4 mL) was added N,N-diisopropylethyl-
amine (0.25 mL, 1.41 mmol), and the mixture was stirred under
an argon atmosphere at room temperature for 30 h. The reac-
tion mixture was diluted with dichloromethane (20 mL),
washed with H2O (10 mL), dried (Na2SO4), and concentrated.
The crude residue was chromatographed on silica gel with
AcOEt–hexane (1:3) to afford compound 6i (230 mg, 93%) as
a light-yellow solid: mp 98–99 ꢀC; 1H NMR (300 MHz,
CDCl3, referenced to TMS) d 1.42 (9H, s), 2.25 (1H, dd,
J=5.1 and 13.8 Hz), 2.78 (1H, dd, J=9.9 and 13.8 Hz), 2.98
(1H, d, J=9.9 Hz), 3.09 (1H, d, J=10.2 Hz), 3.58 (1H, dd,
J=5.1 and 9.2 Hz), 3.66 (3H, s), 3.74 (3H, s), 4.05 (1H, d,
J=14.7 Hz), 4.28 (1H, d, J=14.7 Hz), 5.47 (1H, br s), 7.40
(1H, t, J=6.0 Hz), 7.54 (1H, t, J=7.5 Hz), 7.65 (1H, d,
J=6.9 Hz), 7.84 (1H, d, J=8.1 Hz); 13C NMR (75 MHz,
CDCl3, referenced to central peak of solvent (d=77.0)] d
28.06, 39.34, 52.00, 52.57, 54.52, 61.67, 63.43, 64.13, 79.92,
124.24, 128.05, 130.64, 132.63, 133.49, 149.17, 155.02, 172.22,
173.04. Anal. calcd for C20H27N3O8: C, 54.91; H, 6.22; N,
9.61. Found: C, 54.83; H, 6.14; N, 9.37. The compound 6i
(125 mg, 0.286 mmol) was dissolved in 12 mL of aqueous 6 N
HCl, and the solution was refluxed for 2.5 h. Most of the sol-
vent was removed under reduced pressure, and the residue was
lyophilized to give 9i (103 mg, 94%) as a white solid. Mp 178–
180 ꢀC; 1H NMR (300 MHz, CD3OD, referenced to central
peak of solvent (d=3.31)] d 2.65 (1H, dd, J=9.9 and 14.4 Hz),
References and Notes
1. Nakanishi, S. Science 1992, 258, 597.
2. Nakanishi, S. Neuron 1994, 13, 1031.
3. Reymann, K. G.; Matthies, H. Neurosci.Lett. 1989, 98,
166.
4. Manahan-Vaughan, D.; Reiser, M.; Pin, J.-P.; Wilsch, V.;
Bockaert, J.; Reymann, K. G.; Riedel, G. Neuroscience 1996,
72, 999.
5. Aiba, A.; Cheng, C.; Herrup, K.; Rosenmund, C.; Stevens,
C. F.; Tonegawa, S. Cell 1994, 79, 365.
6. Conn, P. J.; Pin, J.-P. Annu.Rev.Pharmacol.Toxicol. 1997,
37, 205.
7. (a) Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.;
Salhoff, C. R.; Johnson, B. G.; Howe, T.; Alt, C. A.; Rhodes,
G. A.; Robey, R. L.; Griffey, K. R.; Tizzano, J. P.; Kallman,
M. J.; Helton, D. R.; Schoepp, D. D. J.Med.Chem. 1997, 40,
528. (b) Monn, J. A.; Valli, M. J.; Massey, S. M.; Hansen,
M. M.; Kress, T. J.; Wepsiec, J. P.; Harkness, A. R.; Grutsch,
J. L., Jr.; Wright, R. A.; Johnson, B. G.; Andis, S. L.; King-
ston, A. E.; Tomlinson, R.; Lewis, R.; Griffey, K. R.; Tizzano,
J. P.; Schoepp, D. D. J.Med.Chem. 1999, 42, 1027.
8. Schoepp, D. D.; Johnson, B. G.; Salhoff, C. R.; Valli, M. J.;
Desai, M. A.; Burnett, J. P.; Mayne, N. G.; Monn, J. A. Neu-
ropharmacology 1995, 34, 843.

1924
J.K.Mukhopadhyaya et al./ Bioorg.Med.Chem.Lett.11 (2001) 1919–1924
3.16 (1H, dd, J=8.1 and 14.4 Hz), 3.80–3.93 (2H, m), 4.54
(1H, t, J=9.0 Hz), 4.73 (1H, d, J=13.8 Hz), 4.90 (1H, d,
J=13.5 Hz), 7.69 (1H, t, J=7.8 Hz), 7.79 (1H, t, J=7.5 Hz),
7.91 (1H, d, J=7.5 Hz), 8.14 (1H, d, J=7.8 Hz); 13C NMR
(75 MHz, CD3OD, referenced to central peak of solvent
(d=49.15)] d 39.49, 57.14, 61.53, 62.57, 67.53, 126.71, 128.62,
132.26, 134.90, 135.40, 150.69, 170.76, 171.01.
14. Wroblewska, B.; Wroblewski, J. T.; Pshenichkin, S.; Surin,
A.; Sullivan, S.; Neale, J. H. J.Neurochem. 1997, 69, 174.
15. Gomeza, J.; Mary, S.; Brabet, I.; Parmentier, M. L.; Resti-
tuto, S.; Bockaert, J.; Pin, J. P. Mol.Pharmacol. 1996, 50, 923.
16. Valli, M. J.; Schoepp, D. D.; Wright, R. A.; Johnson,
B. G.; Kingston, A. E.; Tomlinson, R.; Monn, J. A. Bioorg.
Med.Chem.Lett. 1998, 8, 1985.
17. Brauner-Osborne, H.; Egebjerg, J.; Nielsen, E. Q.; Mad-
sen, U.; Krogsgaard-Larsen, P. J.Med.Chem. 2000, 43, 2609.
18. Kunishima, N.; Shimada, Y.; Tsuji, Y.; Sato, T.; Yama-
moto, M.; Kumasaka, T.; Nakanishi, S.; Jingami, H.; Mor-
ikawa, K. Nature 2000, 407, 971.
19. Bessis, A.-S.; Bertrand, H.-O.; Galvez, T.; De Colle, C.;
Pin, J.-P.; Acher, F. Protein Sci. 2000, 9, 2200.