Design, synthesis and biological evaluation of novel bicyclo[1.1.1]pentane-based ω-acidic amino acids as glutamate receptors ligands

Article abstract of DOI:10.1016/j.bmc.2008.11.015

A novel series of bicyclo[1.1.1]pentane-based ω-acidic amino acids, including (2S)- and (2R)-3-(3′-carboxybicyclo[1.1.1]pentyl)alanines (8 and 9), (2S)- and (2R)-2-(3′-carboxymethylbicyclo[1.1.1]pentyl)glycines (10 and 11), and (2S)- and (2R)-3-(3′-phosph

Full text of DOI:10.1016/j.bmc.2008.11.015

Bioorganic & Medicinal Chemistry 17 (2009) 242–250
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis and biological evaluation of novel bicyclo[1.1.1]pentane-based
x
-acidic amino acids as glutamate receptors ligands
Rosanna Filosa b, M. Carmela Fulco a, Maura Marinozzi a, Nicola Giacchè a, Antonio Macchiarulo a,
Antonella Peduto b, Antonio Massa c, Paolo de Caprariis b, Christian Thomsen d,
Claus T. Christoffersen d, Roberto Pellicciari a,
*
a Dipartimento di Chimica e Tecnologia del Farmaco, Università di Perugia, Via del Liceo 1, 06123 Perugia, Italy
b Dipartimento di Scienze Farmaceutiche, Università di Salerno, Via Ponte Don Melillo, Variante 11C, 84084 Fisciano, Italy
c Dipartimento di Chimica, Università di Salerno, Via Ponte Don Melillo, Stecca 7, 84084 Fisciano, Italy
d Lundbeck A/S, 9 Ottilavej, DK-2500 Valby, Copenhagen, Denmark
a r t i c l e i n f o
a b s t r a c t
A novel series of bicyclo[1.1.1]pentane-based
x
-acidic amino acids, including (2S)- and (2R)-3-(30-carbo-
Article history:
xybicyclo[1.1.1]pentyl)alanines (8 and 9), (2S)- and (2R)-2-(30-carboxymethylbicyclo[1.1.1]pentyl)gly-
cines (10 and 11), and (2S)- and (2R)-3-(30-phosphonomethylbicyclo[1.1.1]pentyl)glycines (12 and 13),
were synthesized and evaluated as glutamate receptor ligands. Among them, (2R)-3-(30-phospho-
nomethylbicyclo[1.1.1]pentyl)glycine (13) showed relatively high affinity and selectivity at the NMDA
receptor. The results are also discussed in light of pharmacophoric modelling studies of NMDA agonists
and antagonists.
Received 23 September 2008
Revised 30 October 2008
Accepted 3 November 2008
Available online 14 November 2008
Keywords:
Bicyclo[1.1.1]pentane
NMDA receptor
Ó 2008 Elsevier Ltd. All rights reserved.
Metabotropic glutamate receptor
Neurochemistry
Polycycles-bioisosterism
1. Introduction
pharmacophoric groups and was not involved in specific interac-
tions with the receptor. It should be noted, however, that S-CBPG
Bicyclo[1.1.1]pentane is a strained, yet stable cage largely uti-
lized, in virtue of its symmetry, as a valuable unit in the construc-
tion of oligomers with predetermined shape, and polymers and
dendrimers with novel properties.1 The unique properties of the
bicyclo[1.1.1]pentane moiety made it an useful scaffold in the ra-
tional design of small molecules as tool compounds in drug discov-
ery.2 In particular, the ability of the bicyclo[1.1.1]pentane moiety
to keep bridgehead substituents in a coplanar, linear disposition
has been exploited by us for the design of (2S)-2-(30-carboxybicy-
clo[1.1.1]pentyl)glycine (S-CBPG, 1),3 a competitive antagonist of
the subtype 1 of metabotropic glutamate receptors (mGluR1) also
endowed with partial agonist activity on mGluR5 subtype. The idea
behind the design of S-CBPG (1) was the observation that the class
of 4-carboxyphenylglycines (CPGs), firstly reported as group I
mGluRs competitive antagonists,4 founded its pharmacological
profile on the co-linearity imparted by the phenyl ring to the two
(1) is also characterized by a different distance between the phar-
macophoric groups with respect to 4-CPG (2) (5.6 Å vs 6.6 Å,
respectively). With the aim of evaluating the effect of lengthening
the distance between the two pharmacophoric groups in S-CBPG
(1), we carried out the synthesis of (2S)-2-(30-(1H-tetrazol-5-
yl)bicyclo[1.1.1]pentyl)glycine (S-TBPG, 3),5 which proved 2.5-fold
less potent as mGluR1 antagonist than the parent compound 1, and
inactive at mGluR5 subtype. These data indicated that, although
the introduction of the tetrazole ring in place of the
x-carboxylate
moiety increased the pharmacophore distance up to the ‘standard’
one of 4-CPG (2), the substitution was detrimental for the activity
and this could be ascribed to the lower acidity of the tetrazole ring
(pKa = 4.9) with respect to the carboxylate group (pKa = 4.2) or to
the different hydrogen bonding geometry.
To further investigate the role played by the distal acidic moiety
in the class of bicyclopentane derivatives, we recently reported
the synthesis of (2S)-2-(30-phosphonobicyclo[1.1.1]pentyl)glycine
pharmacophoric groups, namely the
x-carboxylate and the glycine
moiety. The ability of S-CBPG (1) to act as group I antagonist sug-
gested that the phenyl ring of CPGs acted as a spacer between
(S-PBPG, 5), the
teric replacement of the
-Glu) or glutamate analogs with the x
x
-phosphonic acid analog of S-CBPG (1).6 The isos-
x
-carboxylic moiety of
L-Glutamic acid
(L
-phosphonic moiety has
indeed represented a productive strategy to obtain different
classes of acidic amino acids endowed with specific activities at
* Corresponding author. Tel.: +39 0755855120; fax: +39 0755855124.
E-mail address: rp@unipg.it (R. Pellicciari).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.11.015

R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
243
ionotropic- and metabotropic glutamate receptors. Thus,
phono amino acids characterized by an appropriate intermediate
chain and by the -configuration at the amino acidic centre, such
as -AP5 (6) and -AP7 (7) have represented the first generation
of potent and selective NMDA competitive antagonists,7 whereas
x-phos-
2. Results and discussion
D
The synthesis of S-CBPA (8) and R-CBPA (9) involved 3-
(methoxycarbonyl)bicyclo[1.1.1]pentane-1-carboxylic acid (14),
synthesized according to the Michl’s procedure,9 as the starting
material (Scheme 1). The one-carbon homologation of the carbox-
ylic moiety of 14 was accomplished by a modified Arndt–Eister
D
D
4-PPG (4), the
x-phosphonate analog of 4-CPG (2), turned out to
be a potent and selective group III mGluRs agonist.8 Accordingly,
S-PBPG (5), characterized by the bioisosteric replacement of the
phenyl ring of 4-PPG (4) by the bicyclo[1.1.1]pentane nucleus
showed moderate potency and selectivity at mGluR4.6
reaction10 involving the formation of
a-diazomethyl ketone 15, ob-
tained by the treatment of the intermediate oxyanhydride with
trimethylsilyldiazomethane,11 then submitted to Wolff rearrange-
ment in aqueous THF in the presence of silver benzoate as catalyst
to give the homologated carboxylic acid 16 (88%). 3-(Methoxycar-
bonyl)bicyclo[1.1.1]pentylacetic acid (16), thus obtained, was con-
verted into the corresponding alcohol 17 by borane reduction
(90%). Dess–Martin oxidation of the primary alcoholic function of
17 led in 76% yield to the corresponding aldehyde 18 which was
H
N N
CO2H
N
N
CO2H
submitted to the condensation with R-
subsequent addition of trimethylsilyl cyanide to the Schiff base
thus formed, afforded the mixture of the two expected -amino ni-
a
-phenylglycinol.12 The
H2N
S
CO2H
H2N
S
CO2H
H2N
CO2H
-CBPG
4-CPG
-TBPG
a
triles, as a major (2S)- and a minor (2R)-component (85:15 by
HPLC).13 Medium pressure liquid chromatography of the reaction
mixture allowed us to obtain 19 and 20, in 35% and 15% yields,
2
3
1
PO3H2
PO3H2
respectively. The a-amino nitrile 19 was submitted to oxidative
cleavage of the chiral auxiliary, acidic hydrolysis and ion exchange
resin chromatography to afford S-CBPA (8) in 73% yield. By an anal-
CO2H
NH2
H2O3P
n
ogous procedure, starting from the minor
a-amino nitrile 20, R-
CBPA (9) was obtained in 85% yield.
H2N
CO2H
-PBPG
H2N
CO2H
n = 1,
n = 3,
D
D
-AP5 6
-AP7 7
The carboxylic acid 16 was also the starting material for the
synthesis of homoCBPG enantiomers 10 and 11, characterized by
S
4-PPG
5
4
the insertion of one methylene unit between the
x-carboxylate
moiety and the bicyclopentane core (Scheme 2). Thus, treatment
of 16 with di-tert-butyl dicarbonate in the presence of DMAP affor-
ded in high yield the mixed diester 21, which was converted by
alkaline hydrolysis into the monoester 22 (87%). Also in this case
the transformation of the carboxylic function of 22 into the corre-
sponding aldehyde 24 was made by a two step-sequence involving
the initial reduction of 22 to the primary alcohol 23 (81%), then
reoxidized with Dess–Martin periodinane (76%).
Starting from the aldehyde 24, the preparation of the title ami-
no acids 10 and 11 was achieved following the synthetic protocol
above described for the synthesis of S-CBPA (8) and R-CBPA (9).
Thus, diastereoselective Strecker reaction, cleavage of the chiral
As a continuation of our work in the design and synthesis of 30-substi-
tuted-2-(bicyclo[1.1.1]pentyl)glycine derivatives and in search for
new excitatory amino acid (EAA) receptor ligands to be used as phar-
macological tools for the characterization of this vast family of recep-
tors, we report herein the synthesis and the preliminary biological
results of (2S)- and (2R)-3-(30-carboxybicyclo[1.1.1]pentyl)alanines
(S-CBPA, 8 and R-CBPA, 9), (2S)- and (2R)-2-(30-carboxymethylbicy-
clo[1.1.1]pentyl)glycines (S-homoCBPG, 10 and R-homoCBPG, 11),
where the carboxylate group is maintained as distal acidic moiety
and the increase of the distance between the pharmacophoric groups
is obtained by homologation of the carbon backbone. The prepara-
tion and the pharmacological evaluation of (2S)- and (2R)-3-(30-
phosphonomethylbicyclo[1.1.1]pentyl)glycines (S-homoPBPG, 12
and R-homoPBPG, 13) are also reported.
auxiliary and final hydrolysis of the
a-amino nitriles 25 and 26,
thus obtained, provided S-homoCBPG (10) and R-homoCBPG (11),
respectively.
The synthesis of the
x-phosphonic acid analogs, S-homo-
PBPG (12) and R-homoPBPG (13), was performed starting from
methyl 3-(hydroxymethyl)bicyclo[1.1.1]pentane-1-carboxylate
(27, Scheme 3).3a Thus, the treatment with bromine and triphenyl-
phosphine, in the presence of imidazole, transformed 27 into the
corresponding bromide 28 (60%), then converted by conventional
Arbuzov reaction conditions into methyl 3-(diethoxyphospho-
ryl)bicyclo[1.1.1]pentane-1-carboxylate (29) in 50% yield. DIBAH
reduction of the carboxylic ester function of 29 in THF at ꢀ60 °C,
led us to obtain the corresponding aldehyde 30 in 53% yield. Start-
ing from 30, according to the three-step sequence already
described, consisting of diastereoselective Strecker reaction, cleav-
HO2C
NH2
H
HO2C
NH2
H
HO2C
HO2C
-CBPA
8
S
R
-CBPA
9
NH2
NH2
H
CO2H
H
CO2H
HO2C
S
HO2C
-homoCBPG
R
-homoCBPG
10
11
age of the chiral auxiliary and final hydrolysis of the two a-amino
nitriles 31 and 32, thus obtained, the preparation of the title com-
pounds S-homoPBPG (12) and R-homoPBPG (13), was completed.
The biological profiles of the synthesized compounds 8–13 were
examined at ionotropic glutamate receptors (iGluRs), NMDA,
AMPA and kainate, and recombinant mGluRs subtypes (mGluR1a,
mGluR2, mGluR4 and mGluR5) in binding experiments performed
as described previously.8,14 As shown in Table 1, S-homoCBPG (10)
NH2
H
NH2
H
H2O3P
H2O3P
CO2H
CO2H
S
-homoPBPG
R
-homoPBPG
12
13

244
R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
O
a
b
N2
MeO2C
CO2H
MeO2C
MeO2C
CO2H
14
16
15
H
N
H
Ph
Ph
MeO2C
MeO2C
MeO2C
N
c
e
R
H
H
NC
NC
OH
OH
R = CH2OH, 17
R = CHO, 18
d
20
19
e
e
HO2C
NH2
H
HO2C
NH2
H
HO2C
HO2C
S-CBPA
R-CBPA
8
9
Scheme 1. Reagents and conditions: (a) i—ClCOOEt, Et3N, THF, ꢀ15 °C, 30’; ii—TMSCHN2, CH3CN, 0 °C, 24 h, 74%; (b) AgOCOPh, Et3N, THF–H2O 4:1, ultrasound, 30’, 88%; (c)
1 M BH3 in THF, THF, ꢀ15 °C, 6 h, 90%; (d) Dess–Martin periodinane, CH2Cl2, 0 °C, 1 h, 76%; (e) i—R-(ꢀ)- -phenylglycinol, MeOH, rt, 8h; ii—TMSCN, 0 °C then rt, 24 h; iii—mplc,
35% (19), 15% (20); (f) i—Pb(OAc)4, CH2Cl2–MeOH, 0 °C, 1 h; ii—6 N HCl, reflux, 24h; iii—Dowex 50WX2-200, 73% (8), 85% (9).
a
was a mGluR1 ligand with moderate affinity (Ki = 59
l
M). It should
To preliminary asses their functional interaction with the
NMDA receptor, the compounds were tested in a [3H]MK801 bind-
ing assay to rat cortical membranes measuring the effects on
potentiation of binding by glutamate/glycine.15 In agreement with
previous observations, glutamate and glycine stimulated
[3H]MK801 binding 3–400% of basal levels. In this assay, com-
pounds 8–13 showed no agonist activity at concentrations up to
1 mM. In contrast, the NMDA antagonist 2(R)-CPP (33), when
be noticed that compound 10 was also able to bind NMDA and
mGluR5, albeit with 2- and 5-fold lower affinities than mGluR1,
respectively.
The most interesting results are related to the affinity of the
compounds 8–13 at the NMDA receptor. In particular, among the
x
-carboxylate derivatives 8–11, R-CBPA (9) and R-homoCBPG
(11) displayed low micromolar affinities at this receptor subtype
and a very good selectivity, being completely inactive at the other
tested in the presence of 30 lM glutamate and 1 lM glycine, inhib-
iGluRs and mGluRs. Both the
x
-phosphonate derivatives, 12 and
ited [3H]MK801 binding with affinity of 103 10 nM. In this assay,
13, showed higher affinities at the NMDA receptor, than the R-
carboxylate ones, still maintaining high degrees of selectivity. In
x
-
only compounds 12 and 13 showed moderate inhibition of
[3H]MK801 binding at higher
lM concentrations (data not shown).
particular, R-homoPBPG (13) resulted the most interesting com-
While these preliminary data suggest that compounds 8–13 have
no agonist activity at the NMDA, they indicate that compounds
12 and 13 may be antagonists at this receptor.
pound of the series, showing a Ki of 0.73
lM at the NMDA receptor
and inactive at the other iGluRs and mGluR subtypes tested.
a
b
c
16
CO2Me
CO2H
R
ButO2C
ButO2C
ButO2C
21
22
R = CH2OH, 23
R = CHO, 24
d
H
N
H
Ph
N
Ph
e
H
H
ButO2C
ButO2C
CN
25
CN
26
OH
OH
f
f
NH2
NH2
H
CO2H
H
CO2H
HO2C
HO2C
S
-homoCBPG
R
-homoCBPG
10
11
Scheme 2. Reagents and conditions: (a) Boc2O, tBuOH, DMAP, rt, 7 h, 88%; (b) NaOHaq, MeOH, rt, 72 h, 87%; (c) 1 M BH3 in THF, THF, ꢀ15 °C, 6 h, 81%; (d) Dess–Martin
periodinane, CH2Cl2, 0 °C, 1 h, 76%; (e) i—R-(ꢀ)- -phenylglycinol, MeOH, rt, 8 h; ii—TMSCN, 0 °C then rt, 24 h; iii—mplc, 30% (25), 15% (26); (f) i—Pb(OAc)4, CH2Cl2–MeOH,
0 °C, 1 h; ii—6 N HCl, reflux, 24 h; iii—Dowex 50WX2-200, 80% (10), 50% (11).
a

R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
245
though a crystallographic study of the NR2 subunit with antago-
nists is not available yet.
Thus, while the pharmacophoric model of NMDA agonists
(Model A) was constructed using a structure-based approach and
b
a
CO2Me
CO2Me
Br
HO
27
28
the experimental folded bioactive conformation of L-Glu as ob-
served in the crystal structure of NR2 (pdb code: 2A5S),20 the phar-
macophoric model of antagonists (Model B) was built using a rigid-
analog approach and the global minimum conformation of the
NMDA competitive antagonist CGS19755 (34, Fig. 2).21
d
c
CO2Me
CHO
Et2O3P
Et2O3P
29
30
H2O3P
H
N
H
N
Ph
H2O3P
H
Ph
N
H
H
Et2O3P
Et2O3P
CN
CN
32
OH
OH
N
CO2H
N
H
CO2H
31
H
e
2(R)-CPP
CGS19755
e
33
34
NH2
NH2
H
Model A was composed of five pharmacophoric points that capture
the interactions observed in the crystal complex between -Glu and
the NR2A subunit of NMDA receptor (pdb code: 2a5s). In particular,
they comprise two hydrogen bond donors located on the -amino
group (D1A, D2A), which are involved in the interactions with the
residues Glu387, Ser486 and Thr488; one negatively charged fea-
L
H
CO2H
H2O3P
H2O3P
CO2H
a
R
-homoPBPG
S
-homoPBPG
13
12
Scheme 3. Reagents and conditions: (a) Br2, Ph3P, imidazole, CH2Cl2, rt, 48 h, 60%;
(b) (EtO)3P, 110 °C, 72 h, 50%; (c) 1.5 M DIBAH in toluene, toluene, ꢀ60 °C, 18 h,
ture located on the
the residue Arg493; and finally, two hydrogen bond acceptor lo-
cated on one oxygen of the -carboxylic group (A1A, A2A) describing
a-carboxylic group (N1A) which interacts with
53%; (d) i—R-(ꢀ)-a-phenylglycinol, MeOH, rt, 3 h; ii—TMSCN, 0 °C then rt, 12 h; iii—
c
mplc, 19% (31), 13% (32); (e) i—Pb(OAc)4, CH2Cl2–MeOH, 0 °C, 1 h; ii—6 N HCl,
reflux, 24 h; iii—Dowex 50WX2-200, 65% (12), 50% (13).
the hydrogen bond formation with residues Ser664 and Thr665.
Model B consisted of four points: one hydrogen bond donor lo-
cated on the
a-amino group of the compounds (DB); three nega-
To further support these observations, we tested the abilities of
compounds 8–13 to fit either a pharmacophoric model of NMDA
agonists (Model A, Fig. 1a) or a pharmacophoric model of NMDA
antagonists (Model B, Fig. 1b).
Briefly, NMDA receptors are obligate heteromers composed of
the glycine-binding NR1, glutamate-binding NR2 and glycine-
binding NR3 subunits.16 Their activation requires the binding of
tively charged groups of which one (N1B) was located on the
a
-
carboxylic group and the others (N2B, N3B) on the two oxygen
atoms of the phosphonic acid moiety. This model is in agreement
with the previously reported pharmacophoric model of competi-
tive antagonists,22 and is consistent with the interaction pattern
described for the NMDA competitive antagonists on the basis of
a mutagenesis study.23 In particular, the latter suggests Arg493,
Ser486, Glu387 and Thr488 as interacting with the amino acidic
moiety of the compounds, and Lys459, Lys462 making salt bridges
with the double negatively charged phosphonic group.
both L-Glu and the co-agonist glycine, plus the relief of the magne-
sium block by membrane depolarization.17,18 According to the ac-
cepted notion concerning the activation mechanism of iGluRs,19
the ligand binding domain (LBD) of the receptor is composed of
two lobes consisting of part of the amino-terminal domain (seg-
ment S1) and the second extracellular loop (segment S2) of the
In Table 2, the number of features matched and the relative fit-
ting score of compounds 8–13 on both models is reported. It is
worth noting that the considerable length between the
a-amino
protein. A folded bioactive conformation of L-Glu and synthetic
acidic moiety and the distal acidic group of compounds 8-13,
typical of NMDA antagonists, hampers the fitting of the bicy-
clo[1.1.1]pentylalanine and -glycine moieties on the agonist phar-
macophoric model (Model A), yielding a null score of fitness.
Conversely, a nice correlation (Fig. 3, r2 = 0.85) is observed
between the binding (pKi) of compounds 8–13 and their fitness
scores on Model B. In particular, the best score obtained for
agonists interacts at both lobes of the LBD promoting the stabiliza-
tion of the functionally active closed conformation of the receptor,
as observed in the crystal structure of the NR2 subunit (pdb code:
2A5S).20 Conversely, an extended bioactive conformation of the
antagonists is proposed to bind at one lobe of the LBD stabilizing
the functionally inactive open conformation of the receptor,
Table 1
Binding affinities values expressed as Ki (lM), for S-CBPA (8), R-CBPA (9), S-homoCBPG (10), R-homoCBPG (11), S-homoPBPG (12) and R-homoPBPG (13)
Compound
NMDA
>300
AMPA
KA
mGluR1a
mGluR5
mGluR2
mGluR4
S-CBPA (8)
R-CBPA (9)
S-homoCBPG (10)
R-homoCBPG (11)
S-homoPBPG (12)
R-homoPBPG (13)
>300
>300
>300
>300
>300
>300
0.17 0.01
>1000
>300
>300
>300
>300
>300
>300
0.092 0.028
>1000
>1000
>300
>1000
>1000
>1000
>1000
nt
nt
6.4 0.5
>1000
>1000
>1000
>1000
>1000
>1000
>1000
1.4 0.2
0.16 0.02
29
124 32
53
4.6 0.8
4
210 30
59 12
>1000
>300
>1000
336 22
293 71
>300
>300
>1000
3
0.73 0.12
0.29 0.09
>300
L
-Glu
-AP4
0.57 0.08
>1000
1.16 0.03
>1000
L
Data are means SEM of 4–6 individual experiments performed in triplicate. For further details, see Section 4. nt, not tested.

246
R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
Figure 1. The pharmacophoric model of NMDA agonists (a) and NMDA antagonists (b). A, hydrogen bonding acceptor site; D, hydrogen bonding donor site; N, negatively
charged group; P, positively charged group. The putative position of interacting residues as observed into the crystal structure of NR2A-NMDA (pdb code: 2A5S) is highlighted
with orange spheres (residues from lobe I), blue spheres (residues from lobe II) and yellow spheres (residues from lobe I).
compounds 12 and 13 in fitting all features of Model B, is in agree-
ment with the lower Ki values shown by these compounds at the
NMDA receptor. On this basis, an antagonistic pharmacological
profile of 12 and 13 at the NMDA receptor may be suggested also
in agreement with the above preliminary data, obtained using
[3H]MK801 binding as a surrogate functional assay.
Although further experimental data, including electrophysiol-
ogy experiments, are required to fully profile the properties of
these compounds, [3H]MK801 binding experiments and pharmaco-
phoric studies suggest a competitive antagonist mode of action for
these compounds at the NMDA receptor.
Finally, R-homoPBPG (13) represents a relatively potent and
selective lead compound for the NMDA receptor and may be an
useful template for generating pharmacological tools to be em-
ployed in the further characterization of this class of glutamate
receptors.
3. Conclusion
Novel competitive NMDA modulators belonging to the class of
bicyclo[1.1.1]pentane-based
sented. S-homoPBPG (12) and R-homoPBPG (13), characterized
by the presence of the phosphonic acid moiety in the -position,
x-acidic amino acids have been pre-
4. Experimental
x
have been shown to have high affinity and selectivity at the NMDA
receptor. In particular, R-homoPBPG (13) displaying an inhibitory
constant in the low micromolar range has the highest affinity in
the series.
All reagents were analytical-grade and purchased from Sigma–
Aldrich (Milan–Italy). Flash chromatography was performed on
Merck silica gel (0.040–0.063 mm). 1H NMR and 13C NMR spectra
were registered on Bruker AC 400 with CDCl3, as solvent, unless
25
20
2
3
α
2
15
34
PO3H2
R
H
R'
10
5
H
H
H
H
R
R'
H
R
H
R'
PO3H2
H
H2O3P
H
0
0
30
60
90
120
150
180
210
240
270
300
330
360
α
Figure 2. Conformational landscape of CGS19755 (34) with its local (black colour coded) and global (red colour coded) minimum conformations.

R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
247
Table 2
Scores of compounds (8–13) in the pharmacophoric Models A and B
Compound
NMDA activity
Model A
Model B
Ki (l
M)
pKi
No of matched sites/No of sites
Fitness score
No of matched sites/No of sites
Fitness score
S-CBPA (8)
R-CBPA (9)
S-HomoCBPG (10)
R-HomoCBPG (11)
S-HomoPBPG (12)
R-HomoPBPG (13)
300
29
124
53
4.6
0.73
ꢀ2.48
ꢀ1.46
ꢀ2.09
ꢀ1.72
ꢀ0.66
0.14
0/5
0/5
0/5
0/5
0/5
0/5
Null
Null
Null
Null
Null
Null
3/4
3/4
3/4
3/4
4/4
4/4
1.886
1.892
1.790
1.816
2.172
2.334
submitted to flash chromatography. Elution with hexane/AcOEt
pKi - Fitness
(80:20) gave 15 (0.520 g, 74%) as a pale-yellow oil; IR (film)
2.500
2.000
1.500
1.000
0.500
2384.6, 2128.2, 1739, 1632.4 cmꢀ1
;
1H NMR: d = 2.27 (s, 6H, 3ꢁ
CH2), 3.64 (s, 3H, CO2CH3), 4.17 ppm (s, 1H, COCHN2); 13C NMR:
11
10
d = 30.1, 37.9, 38.3, 52.3, 53.2, 170.2, 175.2 ppm.
6
7
4.2. 2-(3-Methoxycarbonylbicyclo[1.1.1]pentane)acetic acid (16)
9
8
A solution of silver benzoate (0.47 g, 2.05 mmol) in Et3N
(6.0 mL) was added dropwise to a magnetically stirred solution
of 15 (2.0 g, 10.25 mmol) in THF/water (4:1, 60 mL) and the
resulting mixture was stirred at room temperature for 30 min
under irradiation with ultrasounds. The solvent was partially
removed in vacuo and the residue was diluted with water
(20 mL) and extracted with AcOEt (3ꢁ 10 mL). The aqueous phase
was then extracted with 5% NaHCO3 solution (3ꢁ 30 mL). The
combined basic extracts were acidified until pH 2 and extracted
with AcOEt (4ꢁ 50 ml). The combined organic phases were dried
over anhydrous Na2SO4, filtered and the solvent evaporated in va-
cuo to afford 16 (0.980 g, 88%); 1H NMR: d = 2.18 (s, 6H, 3ꢁ CH2),
2.61 (s, 2H, 2-CH2), 3.72 ppm (s, 3H, s, CO2CH3); 13C NMR:
d = 35.9, 37.2, 38.3, 51.1, 52.3, 170.2, 177.2 ppm; ESIMS m/z
183.2 [Mꢀ1].
f(x) = 0.21x + 2.27
R² = 0.85
0.000
-3
-2.5
-2
-1.5
pKi
-1
-0.5
0
0.5
Figure 3. Correlation plot between the fitness score of Model B (y axis) and the
affinities of compounds 8–13 at the NMDA receptor (x axis).
4.3. Methyl 3-(20-hydroxyethyl)bicyclo[1.1.1]pentane-1-carboxy-
late (17)
otherwise indicated. Chemical shifts are reported in ppm. The
abbreviations are as follows: s, singlet; d, doublet; dd, double dou-
blet; m, multiplet. Optical rotations were recorded on a Jasco Dip-
360 digital polarimeter. GC analyses and MS spectra were carried
out with an HP 6850 gas chromatography equipped with an HP
5975 Mass Selective Detector. ESIMS analysis was carried out on
a Finnigan LCQ Deca ion trap instrument. Microanalyses were car-
ried out on a Carlo Erba 1106 elemental analyzer. The analytical
HPLC analyses were carried out on a Shimadzu (Kyoto, Japan)
Workstation Class LC-10 equipped with a CBM-10A system con-
troller, two LC-10 AD high-pressure binary gradient delivery sys-
tems, an SPD-10A variable wavelength UV/Vis detector, and a
Rheodyne 7725i injector (Rheodyne Inc, Cotati, CA, USA) with a
Boranetetrahydrofurancomplexsolution(1.0 Mintetrahydrofu-
ran, 3.0 mL) was added dropwise to a cold (ꢀ15 °C), magnetically
stirred solution of 16 (0.57 g, 3.11 mmol) in dry THF (3.0 mL) and
the resulting mixture was stirred at ꢀ15 °C for 6 h. The reaction mix-
ture was then neutralized with diluted acetic acid (2 mL) and the or-
ganic solvent partially removed in vacuo. The residue was diluted
with 5% NaHCO3 solution and the solution extracted with ether
(4ꢁ 15 mL). The combined organic phases were dried over anhy-
drous Na2SO4, filtered and the solvent evaporated to give 17
1
(0.47 g, 90%); H NMR: d = 1.82 (t, J = 6.4 Hz, 2H, 10-CH2), 2.10 (s,
6H, 3ꢁ CH2), 3.70 (s, 3H, CO2CH3), 3.72 ppm (t, J = 6.4 Hz 2H, 20-
CH2); 13C NMR: d = 34.1, 38.0, 38.1, 51.8, 52.3, 60.8, 170.2 ppm;
ESIMS m/z 171.2 [M+1].
20 lL stainless steel loop. The optical purity of the final compounds
8–13 were determined by ligand-exchange chromatography.24
4.1. Diazomethyl ketone (15)
4.4. Methyl 3-(2-oxoethyl)bicyclo[1.1.1]pentane-1-carboxylate (18)
Ethyl chloroformate (0.52 g, 3.98 mmol) was added dropwise to a
magnetically stirred solution of 14 (1.06 g, 3.62 mmol) and triethyl-
amine (0.40 g, 3.00 mmol) in dry THF (10 mL) kept at ꢀ15 °C under
vigorous stirring in an argon atmosphere. After 30 min the solid
was filtered off and a solution of (trimethylsilyl)diazomethane
(1.0 g, 1.0 mmol) in dry acetonitrile (40 mL) was directly added to
the solution kept at 0 °C. The resulting mixture was stirred at 0 °C
for 30 min and then stored at ꢀ4 °C for 24 h. After evaporation,
diethyl ether (80 mL) was added and the mixture was washed with
0.5 M aqueous citric acid (30 mL), 5% NaHCO3 solution (30 mL) and
brine (20 mL). The organic layer was dried over anhydrous Na2SO4,
filtered, and the solvent evaporated to afford a residue that was
A solution of 17 (0.46 g, 2.64 mmol) in dry CH2Cl2 (10 mL) was
added dropwise to a cooled (0 °C), magnetically stirred solution of
Dess–Martin periodinane (1.15 g, 2.64 mmol) in dry CH2Cl2
(10 mL). After 30 min at 0 °C, the solution was allowed to warm
to room temperature and 2 N NaOH solution (12 mL) was added.
The aqueous phase was then extracted with ether (2ꢁ 50 mL).
The combined organic phases were washed with brine (10 mL),
dried over anhydrous Na2SO4, filtered and the solvent evaporated
to give a residue that was submitted to flash chromatography. Elu-
tion with hexane/tert-butylmethylether (80:20) afforded 18
(0.340 g, 76%); 1H NMR: d = 2.18 (s, 6H, 3ꢁ CH2), 2.62 (s, 2H,

248
R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
10-CH2), 3.72 (s, 3H, CO2CH3), 9.72 ppm (s, 1H, CHO); 13C NMR:
d = 35.1, 38.3, 45.8, 51.8 52.3, 170.2, 200.1 ppm.
4.9. Methyl 3-(bromomethyl)bicyclo[1.1.1]pentane-1-carboxylate
(28)
4.5. Methyl 3-(tert-butoxycarbonylmethyl)bicyclo[1.1.1]pen-
tane-1-carboxylate (21)
Bromine (6.0 g, 37.5 mmol) was added dropwise to a magneti-
cally stirred solution of triphenylphosphine (23.7 g, 90 mmol) in
dry CH2Cl2 (118 mL) kept at room temperature under an argon
atmosphere. After 10 min, imidazole (3.57 g, 52 mmol) was added.
A solution of 27 (3.3 g, 21 mmol) in dry CH2Cl2 (115 mL) was then
added over the course of 20 min and the resulting mixture was
stirred for 48 h at room temperature. The mixture was washed
with water (2ꢁ 100 mL) and brine (2ꢁ 100 mL), dried over anhy-
drous Na2SO4, filtered and the solvent evaporated to afford a resi-
due that was submitted to flash chromatography. Elution with
light petroleum/AcOEt (95:5) afforded 28 (2.8 g, 60%); 1H NMR:
d = 1.96 (s, 6H, 3ꢁ CH2), 3.37 (s, 2H, CH2Br), 3.61 ppm (s, 3H,
CO2CH3); 13C NMR: d = 32.50, 36.69, 38.98, 50.82, 51.22, 169.48
ppm; GC-MS m/z (relative intensity): 219 [M+](<1), 139 (26), 107
(27), 79 (100), 59 (20).
4-(Dimethylamino)pyridine (0.08 g, 0.64 mmol) and di-tert-bu-
tyl dicarbonate (0.95 g, 4.35 mmol) were added to a magnetically
stirred solution of 16 (0.40 g, 2.13 mmol) in dry tert-butanol
(10 mL). After 7 h, the solvent was partially removed in vacuo,
the residue was diluted with ether (20 mL) and the solution
washed with H2O (10 mL). The organic phase was dried over anhy-
drous Na2SO4, filtered, and the solvent evaporated in vacuo to give
a residue that was submitted to flash chromatography. Elution
with hexane/AcOEt (90:10) afforded 21 (0.45 g, 88%); 1H NMR:
d = 1.40 (s, 9H, C(CH3)3), 2.05 (s, 6H, 3ꢁ CH2), 2.40 (s, 2H,
CH2CO2C(CH3)3), 3.65 ppm (s, 3H, CO2CH3); 13C NMR: d = 28.2,
35.9, 38.2, 38.4, 51.1, 52.3, 78.84, 170.2 ppm.
4.6. 3-(tert-Butoxycarbonylmethyl)bicyclo[1.1.1]pentane-1-
carboxylic acid (22)
4.10. Methyl 3-[(diethoxyphosphoryl)methyl]bicyclo[1.1.1]-
pentane-1-carboxylate (29)
A solution of sodium hydroxide (0.072 g, 1.8 mmol) in water
(2.0 mL) was added dropwise to a magnetically stirred solution
of 21 (0.43 g, 1.77 mmol) in MeOH (4.0 mL). After 72 h, the solvent
was partially removed, the residue diluted with water (15 mL) and
extracted with ether (3ꢁ 10 mL). The aqueous layer was acidified
to pH 3 with 10% citric acid solution and extracted with ethyl ace-
tate (4ꢁ 20 ml). The combined organic phases were dried over
anhydrous Na2SO4, filtered and the solvent removed in vacuo to af-
ford 22 (0.35 g, 87%); 1H NMR: d = 1.40 (s, 9H, C(CH3)3), 2.05 (s, 6H,
3ꢁ CH2), 2.42 ppm (2H, CH2CO2C(CH3)3); 13C NMR: d = 25.8, 35.9,
36.15, 36.53, 50.5,78.8, 168.31, 173.6 ppm; ESIMS m/z 225.0
[Mꢀ1].
A magnetically stirred mixture of 28 (2.7 g, 12.3 mmol) and tri-
ethylphosphite (25 mL) was heated at 110 °C for 72 h. Excess of tri-
ethylphosphite was distilled off and the residue purified by flash
chromatography. Elution with AcOEt/light petroleum (80:20)
afforded 29 (1.7 g, 50%); 1H NMR: d = 1.24 (t, J = 7. 0 Hz, 6H, 2ꢁ
PO3CH2CH3), 2.01 (m, 8H, 3ꢁ CH2 and CH2 PO3CH2CH3), 3.59 (s,
3H, CO2CH3), 4.10 ppm (m, 4H, 2ꢁ PO3CH2CH3); 13C NMR:
d = 16.42 (d, JCP = 6.1 Hz), 28.18 (d, JCP = 139 Hz), 34.08 (d,
JCP = 8.0 Hz), 38.26, 51.55, 53.34 (d, JCP = 7.7 Hz), 61.38 (d,
JCP = 6.3 Hz), 170.00 ppm. 31P NMR (81 MHz): d = 29.1 ppm; GC-
MS m/z (relative intensity): 277 [M+](3), 245 (16), 216 (47), 188
(100), 160 (63), 139 (17), 107 (32), 79 (47).
4.7. tert-Butyl (3-(hydroxymethyl)bicyclo[1.1.1]pentyl)acetate
(23)
4.11. Diethyl (3-formylbicyclo[1.1.1]pent-1-yl)methylphosphonate
(30)
Borane tetrahydrofuran complex solution (1.0 M in THF, 2.0 mL)
was added dropwise to a cold (ꢀ15 °C), magnetically stirred solu-
tion of 22 (0.45 g, 1.99 mmol) in dry THF (3.0 mL) and the resulting
mixture was stirred at ꢀ15 °C for 6 h. The reaction mixture was
then neutralized with diluted acetic acid (2 mL) and the organic
solvent partially removed in vacuo. The residue was diluted with
5% NaHCO3 solution and the solution extracted with ether
(4ꢁ 15 mL). The combined organic phases were dried over anhy-
drous Na2SO4, filtered and the solvent evaporated to give 23
(0.34 g, 81%); 1H NMR: d = 1.41 (s, 9H, C(CH3)3), 1.67 (s, 6H, 3x
CH2), 2.42 (s, 2H, CH2CO2C(CH3)3), 3.63 ppm (s, 2H, CH2OH); 13C
NMR: d = 26.2, 34.5, 37.2, 38.4, 47.7, 61.2, 78.64, 168.9 ppm.
Precooled (ꢀ78 °C) diisobutylaluminium hydride (4.5 mL,
6.75 mmol, 1.5 M in toluene) was added dropwise, within 20 min,
to a cooled (ꢀ78 °C), magnetically stirred solution of 29 (1.7 g,
6.15 mmol) in dry toluene (68 mL) kept under an argon atmosphere.
After the end of the addition, the reaction mixture was warmed to
ꢀ55 °C. After 4 h, because the reaction was not complete, additional
diisobutylaluminium hydride (2.3 mL, 3.37 mmol, 1.5 M in toluene)
was added, and the mixture allowed to stir at ꢀ55 °C for 18 h. The
reaction was quenched by the addition of methanol (8 mL). HCl
(1 M, 30 mL) was then added and the mixture was extracted with
AcOEt (4ꢁ 40 mL). The combined organic phases were dried over
anhydrous Na2SO4, filtered and the solvent evaporated to afford a
residue that was submitted to flash chromatography. Elution with
AcOEt/light petroleum (90:10) afforded 30 (0.80 g, 53%); 1H NMR:
d = 1.24 (t, J = 7. 0 Hz, 6H, 2ꢁ PO3CH2CH3), 2.01 (m, 8H, 3ꢁ CH2 and
CH2 PO3CH2CH3), 3.59 (s, 3H, CO2CH3), 4.01 (m, 4H, 2ꢁ PO3CH2CH3),
9.46 ppm (s, 1H, CHO); 13C NMR: d = 16.45 (d, JCP = 6.0 Hz), 28.50 (d,
JCP = 138.7 Hz), 35.08, 44.75, 52.16 (d, JCP = 7.8 Hz), 61.53 (d,
JCP = 6.4 Hz), 198.43 ppm. 31P NMR (81 MHz) d = 28.9 ppm; GC-MS
m/z (relative intensity): 245 [M+](3), 217 (25), 189 (26), 161 (68),
109 (58), 92 (80), 79 (100).
4.8. tert-Butyl (3-formylbicyclo[1.1.1]pentane)acetate (24)
A solution of 23 (0.32 g, 1.50 mmol) in dry CH2Cl2 (10 mL) was
added dropwise to a cooled (0 °C), magnetically stirred solution of
Dess–Martin periodinane (0.65 g, 1.49 mmol) in dry CH2Cl2
(10 mL). After 30 min at 0 °C, the solution was allowed to warm
to room temperature and 2 N NaOH solution (12 mL) was added.
The aqueous phase was then extracted with ether (2ꢁ 50 mL).
The combined organic phases were washed with brine (10 mL),
dried over anhydrous Na2SO4, filtered and the solvent evaporated
to give a residue that was submitted to flash chromatography. Elu-
tion with hexane/tert-butylmethylether (50:50) afforded 24
(0.24 g, 76%); 1H NMR: d = 1.41 (s, 9H, C(CH3)3), 2.02 (s, 6H, 3ꢁ
CH2), 2.42 (s, 2H, CH2CO2C(CH3)3), 9.54 (s, 1H, CHO); 13C NMR:
d = 25.5, 34.8, 36.7, 42.5, 48.6, 78.64, 168.9, 196.6 ppm.
4.12. General procedure for the Strecker synthesis to give
substituted a-amino nitriles
(R)-2-Phenylglycinol (1.0 mmol) was added to a solution of the
aldehyde (1.0 mmol) in methanol (7 mL), and the resulting

R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
249
solution was magnetically stirred at room temperature for 8 h.
After cooling at 0 °C, TMSCN (2.0 mmol) was added and the result-
ing mixture was stirred for 12 h at room temperature. Evaporation
of the solvent gave a residue that was submitted to medium pres-
28.29 (d, JCP = 138.8 Hz), 33.94 (d, JCP = 7.8 Hz), 39.62, 49.58,
50.77 (d, JCP = 7.6 Hz), 61.57 (d, JCP = 6.0 Hz), 63.28, 66.46, 118.48,
127.48, 128.23, 128.81, 139.52 ppm. 31P NMR (81 MHz):
d = 29.1 ppm; ½a 20
= ꢀ24.0 (c 1.3, CHCl ).
ꢂ
D
3
sure chromatography to afford the corresponding N-substituted a-
amino nitriles.
4.19. General procedure for the oxidative cleavage and
hydrolysis of N-substituted -amino nitriles
a
4.13. N-[(R)-2-Phenylglycinyl]-a-amino nitrile (19)
Lead(IV) acetate (0.45 mmol) was added to a cold (0 °C),
magnetically stirred solution of the -amino nitrile (0.38 mmol)
Light petroleum/AcOEt (70:30); 35% yield; 1H NMR: d = 2.01 (m,
2H, 1-CH2), 2.05 (s, 6H, 3ꢁ CH2), 3.27 (t, J = 6.8 Hz, 1H, CHCN), 3.59
(t, J = 9.9 Hz, 1H, CHPh), 3.70 (s, 3H, CO2CH3), 3.80 (dd, J = 4.1 and
9.6 Hz, 1H, CHaOH), 4.11 (dd, J = 4.1 and 9.6 Hz, 1H, CHbOH),
7.37 ppm (m, 5H, aromatics); 13C NMR: d = 35.5, 37.4, 38.2, 45.9,
52.0, 52.9, 63.4, 67.4, 120.1, 128.1, 128.4, 128.9, 138.6,
a
in dry MeOH/CH2Cl2 (1:2, 10 mL). After 10 min, the cooling bath
was removed, and stirring continued for 1 h. Water (5 mL) was
then added, and the resulting mixture was filtered with the aid
of Celite. After evaporation of the solvent, the residue was
heated at 95 °C in 6 N HCl (10 mL) for 18–24 h. The reaction
mixture was evaporated to dryness and the residue was
submitted to ion-exchange resin chromatography (Dowex
50WX2-200).
170.3 ppm; ½a D
ꢂ
= ꢀ107.0 (c 1, CHCl3); ESIMS m/z 315.4 [M+1].
20
4.14. N-[(R)-2-Phenylglycinyl]-a-amino nitrile (20)
Light petroleum/AcOEt (70:30); 15% yield; 1H NMR: d = 2.01 (m,
2H, 1-CH2), 2.05 (s, 6H, 3ꢁ CH2), 3.75-3.8 (m, 2H, CHPh and
CHaOH), 3.59 (t, J = 9.9 Hz, 1H, CHCN), 3.70 (s, 3H, CO2CH3), 3.80
(m, 1H, CHbOH), 7.37 ppm (m, 5H, aromatics); 13C NMR: d = 35.5,
37.2, 38.4, 47.2, 52.0, 52.8, 63.2, 66.9, 120.5, 127.8, 128.2, 129.2,
4.20. (2S)-3-(30-Carboxybicyclo[1.1.1]pentyl)alanine (S-CBPA, 8)
Pyridine (10%); 73% yield; 1H NMR (D2O): d = 2.04 (s, 6H, 3ꢁ CH2),
2.14 (m, 2H, 3-CH2), 3.93 ppm (t, J = 6.8 Hz, 1H, 2-CH); 13CNMR
(D2O): d = 29.67, 33.53, 35.66, 49.77, 170.77, 172.70 ppm; 96.0%
140.1, 170.0 ppm; ½a D
ꢂ
= ꢀ31.6 (c 1, CHCl ); ESIMS m/z 315.4
ee; ½a D
ꢂ
= +16.2 (c 1, D O); Anal. calcd for C H13NO4: C, 54.26; H,
20
2 9
20
3
[M+1].
6.58; N, 7.03. Found: C, 54.28; H, 6.57; N, 7.04.
4.15. N-[(R)-2-Phenylglycinyl]-
a
-amino nitrile (25)
4.21. (2R)-3-(30-Carboxybicyclo[1.1.1]pentyl)alanine (R-CBPA, 9)
Light petroleum/EtOAc (70:30); 30% yield; 1H NMR: d = 1.48 (s,
9H, C(CH3)3), 1.83 (m, 6H, 3ꢁ CH2), 2.48 (s, 2H, CH2) 3.4 (s, 1H,
CHCN), 3.59 (t, J = 10.3 Hz, 1H, CHPh), 3.80 (dd, J = 4.1 and
10.1 Hz, 1H, CHaOH), 4.35 (dd, J = 4.1 and 10.1 Hz, 1H, CHbOH),
7.37 ppm (m, 5H, aromatics); 13C NMR: d = 26.2, 34.2, 36.6, 37.1,
47.1, 47.6, 61.1, 65.5, 78.6, 116.5, 125.6, 126.3, 126.9, 136.3,
Pyridine (10%); 85% yield; 96.5% ee; ½a D
ꢂ
= ꢀ16.7 (c 1, D2O);
20
Anal. calcd for C9H13NO4: C, 54.26; H, 6.58; N, 7.03. Found: C,
54.27; H, 6.59; N, 7.05.
4.22. (2S)-2-(30-Carboxymethylbicyclo[1.1.1]pentyl)glycine (S-
homoCBPG, 10)
168.4 ppm; ½a D
ꢂ
= ꢀ110.3 (c 1, CHCl3); ESIMS m/z 357.5 [M+1].
20
Pyridine (10%); 80% yield; 1H NMR (D2O): d = 1.80 (s, 6H, 3ꢁ
CH2), 2.50 (s, 2H, s, CH2CO2H); 3.93 ppm (s, 1H, 2-CH); 13C NMR
(D2O): d = 35.11, 36.78, 36.81, 50.44, 54.08, 170.89, 176.61 ppm;
4.16. N-[(R)-2-Phenylglycinyl]-a-amino nitrile (26)
Light petroleum/EtOAc(70:30); 15% yield; 1H NMR: d = 1.45 (s,
9H, C(CH3)3), 1.83 (m, 6H, 3ꢁ CH2), 2.46 (s, 2H, CH2), 3.75–3.8
(m, 2H, CHPh and CHaOH), 3.77 (s, 1H, CHCN), 3.9 (dd, J = 4.8 and
7.3 Hz, 1H, CHaOH), 7.37 ppm (m, 5H, aromatics); 13CNMR:
d = 25.4, 34.0, 35.6, 37.2, 47.8, 47.6, 61.1, 64.3, 78.7, 116.4, 125.6,
96% ee; ½a D
ꢂ
= +14.5 (c 1, D2O); Anal. calcd for C9H13NO4: C,
54.26; H, 6.58; N, 7.03. Found: C, 54.28; H, 6.59; N, 7.06.
20
4.23. (2R)-2-(30-Carboxymethylbicyclo[1.1.1]pentyl)glycine (R-
homoCBPG, 11)
126.3, 126.9, 136.7, 168.14 ppm; ½a D
ꢂ
= ꢀ34.4 (c 1, CHCl3); ESIMS
20
m/z 357.5 [M+1].
Pyridine (10%); 50% yield; 90% ee; ½a D
ꢂ
= ꢀ8.5 (c 1, D2O); Anal.
20
calcd for C9H13NO4: C, 54.26; H, 6.58; N, 7.03. Found: C, 54.27; H,
6.60; N, 7.04.
4.17. N-[(R)-2-Phenylglycinyl]-a-amino nitrile (31)
EtOAc/MeOH (98:2); 19% yield; 1H NMR: d = 1.24 (t, J = 7. 0 Hz,
6H, 2ꢁ PO3CH2CH3), 1.77 (s, 6H, 3ꢁ CH2), 1.98 (d, J = 18.8 Hz,
CH2PO3CH2CH3), 3.28 (s, 1H, CHCN), 3.47 (m, 1H, CHaOH), 3.68
(dd, J = 3.6 and 10.8 Hz, 1H, CHbOH), 4.00 (m, 5 H, 2ꢁ PO3CH2CH3
and CHPh), 7.25 ppm (m, 5H, aromatics); 13C NMR: d = 16.48 (d,
JCP = 6.0 Hz), 28.33 (d, JCP = 138.5 Hz), 33.09 (d, JCP = 7.9 Hz),
39.14, 48.94, 50.50 (d, JCP = 7.6 Hz), 61.54 (d, JCP = 6.4 Hz), 63.02,
67.34, 118.40, 127.57, 128.15, 128.80, 138.49 ppm; 31P NMR
4.24. (2S)-(30-Phosphonomethylbicyclo[1.1.1]pentyl)glycine (S-
homoPBPG, 12)
Pyridine (10%); 65% yield; 1H NMR (D2O): d = 1.64 (s, 6H, 3ꢁ
CH2), 1.77 (d, J = 12.0 Hz, CH2PO3H2), 3.73 ppm (s, 1H, 2-CH); 13C
NMR (D2O): d = 29.46 (d, JCP = 131.4 Hz), 33.58 (d, JCP = 7.0 Hz),
37.03, 50.66 (d, JCP = 7.5 Hz), 54.18, 171.15 ppm. 31P NMR
(81 MHz) d = 25.1 ppm; 96% ee; ½a 20
ꢂ
= +9.5 (c 0.75, H2O); Anal.
calcd for C8H14NO5P: C, 40.86; H, 6.00; N, 5.96. Found: C, 40.88;
H, 6.03; N, 5.97.
D
(81 MHz): d = 29.3 ppm; ½a 20
ꢂ
= ꢀ68.8 (c 0.9, CHCl ).
D
3
4.18. N-[(R)-2-Phenylglycinyl]-a-amino nitrile (32)
4.25. (2R)-(30-Phosphonomethylbicyclo[1.1.1]pentyl)glycine (R-
homoPBPG, 13)
EtOAc/MeOH (97:3); 13% yield; 1H NMR: d = 1.24 (t, J = 7. 0 Hz,
6H, 2ꢁ PO3CH2CH3), 1.81 (s, 6H, 3ꢁ CH2), 1.95 (d, J = 18.8 Hz,
CH2PO3CH2CH3), 3.46-3.69 (m, 3H, CHCN, CHaOH and CHPh), 3.89
Pyridine (10%); 50% yield; 95% ee; ½a 20
= ꢀ9.0 (c 0.75, H O);
ꢂ
D
2
(dd, J = 5.2 and 8.8 Hz, 1H, CHbOH), 4.00 (m, 4 H, 2ꢁ PO3CH2CH3),
Anal. calcd for C8H14NO5P: C, 40.86; H, 6.00; N, 5.96. Found: C,
40.87; H, 6.02; N, 5.97.
13
7.26 ppm (m, 5H, aromatics);
C NMR: d = 16.48 (d, JCP = 6.0 Hz),

250
R. Filosa et al. / Bioorg. Med. Chem. 17 (2009) 242–250
5. Molecular modelling
References and notes
1. (a) Bunz, U.; Polborn, K.; Wagner, H. U.; Szeimies, G. Chem. Ber. 1988, 121,
1785–1790; (b) Schlueter, A. D.; Bothe, H.; Gosau, J. M. Makrom. Chem. 1991,
192, 2497–2519; (c) Ramireddy, C.; Reddy, V. S.; Munk, P.; Wu, C. N.
Macromolecules 1991, 24, 1387–1391.
2. (a) Barbachyn, M. R.; Hutchinson, D. K.; Toops, D. S.; Reid, R. J.; Zurenko,
G. E.; Yagi, B. H.; Schaadt, R. D.; Allison, J. W. Bioorg. Med. Chem. Lett.
1993, 3, 671–676; (b) Pritz, S.; Pätzel, M.; Szeimies, G.; Dathe, M.; Bienert,
M. Org. Biomol. Chem. 2007, 5, 1789–1794; (c) Mikhailiuk, P. K.; Afonin, S.;
Chernega, A. N.; Rusanov, E. B.; Platonov, M. O.; Dubinina, G. G.; Berditsch,
M.; Ulrich, A. S.; Komarov, I. V. Angew. Chem., Int. Ed. 2006, 45, 5659–
5661.
Pharmacophoric studies were performed using the program
Phase v.2.0 as implemented in the software package Schrödinger
suite 2006.25 All ligands were constructed, prepared and energy re-
fined with the program LigPrep v. 2.0,26 assigning the correct chi-
rality and considering all possible ionic states at pH 7 ( 2). The
conformational analysis of compound 34 was carried out using
the coordinate scan module of MacroModel as implemented in
the Schrödinger suite and the following setting: conformational
window 360°; torsional angle increment 10°; forcefield OPLS
2005; gradient convergence threshold of energy minimization
0.05; implicit consideration of water solvent.
3. (a) Pellicciari, R.; Raimondo, M.; Marinozzi, M.; Natalini, B.; Costantino, G.;
Thomsen, C. J. Med. Chem. 1996, 39, 2874–2876; (b) Mannaioni, G.; Attucci, S.;
Missanelli, A.; Pellicciari, R.; Corradetti, R.; Moroni, F. Neuropharmacology 1999,
38, 917–926.
The pharmacophoric searches were carried out using Phase and
its default parameters. The pharmacophoric model of NMDA ago-
nists (Model A) was constructed using the structural information
4. Eaton, S. A.; Jane, D. E.; Jones, P. L.; Porter, R. H.; Pook, P. C.; Sunter, D. C.;
Udvarhelyi, P. M.; Roberts, P. T.; Watkins, J. C. Eur. J. Pharmacol. 1993, 244, 195–
197.
5. Costantino, G.; Maltoni, K.; Marinozzi, M.; Camaioni, E.; Prezeau, L.; Pin, J.-P.;
Pellicciari, R. Bioorg. Med. Chem. 2001, 9, 221–227.
obtained from the inspection of the binding mode of L-Glu ob-
6. Filosa, R.; Marinozzi, M.; Costantino, G.; Hermit, M. B.; Thomsen, C.; Pellicciari,
R. Bioorg. Med. Chem. Lett. 2006, 14, 3811–3817.
7. (a) Watkins, J. C.; Evans, R. H. Neurosci. Lett. 1981, 21, 77–81; (b)
Olverman, H. J.; Jones, A. W.; Mewett, K. N.; Watkins, J. C. Neuroscience
1988, 26, 17–31.
8. (a) Gasparini, F.; Inderbitzin, W.; Francotte, E.; Lecis, G.; Richert, P.; Gragic, Z.;
Kuhn, R.; Flor, P. J. Bioorg. Med. Chem. Lett. 2000, 10, 1241–1244; (b) Sabelhaus,
C. F.; Schroder, U. H.; Breder, J.; Henrich-Noack, P.; Reymann, K. G. Br. J.
Pharmacol. 2000, 131, 655–658; (c) Palucha, A.; Tatarczynska, E.; Branski, P.;
Szewczyk, B.; Wieronka, J. M.; Klak, K.; Chojnacka-Wojcik, E.; Nowak, G.; Pilc, A.
Neuropharmacology 2004, 46, 151–159.
served in the crystal structure of the NR2 subunit of the NMDA
receptor (pdb code: 2A5S) and according to the interaction model
proposed by Laube and coworkers.23 Conversely, a rigid-analog ap-
proach based on the global minimum conformation of compound
34 was used for the construction of the pharmacophoric model
of NMDA antagonists (Model B).22
For the evaluation of our compounds 8–13 on both agonist and
antagonist pharmacophoric models, we used the program Phase
with the following specific setting: the tolerance of matching a fea-
ture was set to 1.0 Å and 2.0 Å, respectively, for the agonist (Model
A) and antagonist models (Model B); the minimum number of re-
quired features to be fitted by a compound was set to 4 of 5 and 3
of 4, respectively, for Model A and Model B. Furthermore, while in
Model A two hydrogen bond acceptor features (A1A and A2A, Fig. 1)
and one negative charge feature (N1A, Fig. 1) were defined as obli-
9. Kaszynski, P.; McMurdie, N. D.; Michl, J. J. Org. Chem. 1991, 56,
307–316.
10. Pellicciari, R.; Natalini, B.; Marinozzi, M. Synth. Commun. 1988, 18(14), 1707–
1710.
11. Shioiri, T.; Aoyama, T. In Handbook of Reagents for Organic Synthesis, Reagents,
Auxiliaries and Catalysts for C–C Bond Formation; Coates, R. M., Denmark, S. E.,
Eds.; Wiley & Sons LTD, 1999; pp 662–665.
12. Chakraborty, T. K.; Hussian, K. A.; Reddy, G. V. Tetrahedron 1995, 51, 9179–
9190.
gate matching, in Model B the
a-aminoacid feature (N1B and DB)
13. Pellicciari, R.; Marinozzi, M.; Natalini, B.; Costantino, G.; Luneia, R.; Giorgi, G.;
Moroni, F.; Thomsen, C. J. Med. Chem. 1996, 39, 2259–2269.
14. Pellicciari, R.; Marinozzi, M.; Macchiarulo, A.; Fulco, M. C.; Gafarova, J.;
Serpi, M.; Giorgi, G.; Nielsen, S.; Thomsen, C. J. Med. Chem. 2007, 5019,
4630–4641.
was defined as obligate matching.
6. Pharmacology
15. Urwyler, S.; Laurie, D.; Lowe, D. A.; Meier, C. L.; Müller, W. Neuropharmacology
1996, 35(6), 643–654.
Membranes from BHK cells expressing subtypes of mGluRs
(mGluR1, mGluR2, mGluR4 and mGluR5) were used for receptor
binding experiments as previously described.8,14 [3H]Quisqualate
binding were used for mGluR1 and mGluR5, [3H]LY341495 binding
16. Paoletti, P.; Neyton, J. Curr. Opin. Pharmacol. 2007, 7, 39–47.
17. Benveniste, M.; Mayer, M. L. Biophys. J. 1991, 59, 560–573.
18. Clements, J. D.; Westbrook, G. L. Neuron 1991, 7, 605–613.
19. Oswald, R. E.; Ahmed, A.; Fenwick, M. K.; Loh, A. P. Curr. Drug Targets 2007, 8,
573–582.
20. Furukawa, H.; Singh, S. K.; Mancusso, R.; Gouaux, E. Nature 2005, 438, 185–
192.
21. Lehmann, J.; Hutchison, A. J.; McPherson, S. E.; Mondadori, C.; Schmutz, M.;
Sinton, C. M.; Tsai, C.; Murphy, D. E.; Steel, D. J.; Williams, M. J. Pharmacol. Exp.
Ther. 1988, 246, 65–75.
for mGluR2 and [3H]
L
-AP4 binding for mGluR4. Affinities for native
AMPA, kainate and NMDA receptors were determined using 5 nM
[3H]AMPA, 5 nM [3H]kainic acid and 2 nM [3H]CGP39653 ([3H]
D,L-
(E)-2-amino-4-propyl-5-phosphono-3-pentenoic acid) as previ-
ously described.8,14
22. Whitten, J. P.; Harrison, B. L.; Weintraub, H. J.; McDonald, I. A. J. Med. Chem.
1992, 35, 1509–1514.
23. Laube, B.; Schemm, R.; Betz, H. Neuropharmacology 2004, 47(7),
994–1007.
24. Natalini, B.; Sardella, R.; Macchiarulo, A.; Pellicciari, R. J. Sep. Sci. 2008, 31, 696–
704.
25. Phase, version 2.0, Schrödinger, LLC: New York, NY, 2006.
26. LigPrep, version 2.0, Schrödinger, LLC: New York, NY, 2006.
Acknowledgment
This work was supported by the Italian Ministry of Research
(MIUR-PRIN 2006).