Synthesis of a series of γ-amino alcohols comprising an N-methyl isoindoline moiety and their evaluation as NMDA receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2011.07.119

We report a series of new stereoisomeric γ-amino alcohols comprising an N-methyl isoindoline moiety as ligands for the ifenprodil binding site of the NMDA receptor. Among the four series of stereoisomers, 8a-c, 9a-c, 10a-c, and 11a-c, synthesised, the hig

Full text of DOI:10.1016/j.bmcl.2011.07.119

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5795–5799
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of a series of c-amino alcohols comprising an N-methyl
isoindoline moiety and their evaluation as NMDA receptor antagonists
Andreas Müller a, Georg Höfner a, Thejavathi Renukappa-Gutke a, Chris G. Parsons b, Klaus T. Wanner a,
⇑
a Department of Pharmacy, Ludwig-Maximilians-University Munich, Butenandtstr. 5-13, D-81377 Munich, Germany
b Merz Pharmaceuticals GmbH, Eckenheimer Landstr. 100-104, D-60318 Frankfurt am Main, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
We report a series of new stereoisomeric
c
-amino alcohols comprising an N-methyl isoindoline moiety as
Received 1 July 2011
Revised 29 July 2011
Accepted 31 July 2011
Available online 6 August 2011
ligands for the ifenprodil binding site of the NMDA receptor. Among the four series of stereoisomers,
8a–c, 9a–c, 10a–c, and 11a–c, synthesised, the highest potencies and NMDA-NR2B subtype selectivity
was found for the methyl derivative 11a and the chloro derivative 11c, both possessing the [1S,10S] con-
figuration. However, additional moderate potency of 11a and 11c at the hERG channel with values of
2.6 2.4% and 1.6 2.0%, respectively, rendered them unsuitable for medical use.
Ó 2011 Elsevier Ltd. All rights reserved.
Dedicated to Prof. Eberhard Reimann with
best wishes on occasion of his 75th
birthday.
Keywords:
NMDA
Ifenprodil
Isoindoline
Amino alcohol
Biological activity
Antagonists acting at different recognition sites of NMDA recep-
tors are potential therapeutics for numerous neurological and psy-
chiatric disorders and, for example, exert neuroprotective effects in
various ischemia models.1–4 Among these, substances binding to
the ifenprodil (or polyamine) site and thereby selectively antagon-
ising NR2B subunit containing NMDA receptors are of special inter-
est as they may be suited to reduce the psychostimulant, amnesic
and neurotoxic effects observed for many other types of NMDA
receptor antagonists.5–9 The first such antagonists to be found
were erythro-ifenprodil,10,11 its analogs eliprodil (2)12 and Ro 25-
6981 (3)13 leading to the development of numerous further ligands
for that site (Scheme 1). The presence of two substituted aromatic
rings and a piperidine or a pyrrolidine unit is characteristic for
most of these compounds, although examples exist where the ami-
no functionality is not part of a ring (see Scheme 1).14 In addition,
the amino alcohol moiety present in ifenprodil and eliprodil is not
found in all substances acting as antagonists at the ifenprodil bind-
ing site. This is exemplified by potent NR2B selective antagonists 4
compounds.16–18 For example, threo-ifenprodil binds more selec-
tively to NMDA-receptors than its erythro-isomer which is 1-
a
selective.17,18 Or in the case of eliprodil, according to recent results
from our group, the (R)-enantiomer exhibits a distinctly higher
affinity for NMDA receptors containing NR2B subunits than the
(S)-form.18 A similar enantioselectivity of binding to the NMDA
receptor has been reported for (+)-CP-101.606 (6) and its enantio-
mer (ꢀ)-CP-101.581, the former being far more potent than the
latter.17
As part of a research project aiming at the stereoselective syn-
thesis of nitrogen heterocycles having a 2-phenyl-2-hydroxylethyl
substituent in the a-position to the nitrogen we prepared a series
of tetrahydroisoquinoline derivatives.19 These were revealed to act
as antagonists at the ifenprodil binding site of the NMDA receptor
with compound 7 (Scheme 1) displaying the highest affinity of the
compounds studied.20 Unfortunately, it appeared that the isoquin-
oline derivative 7 also effectively binds to the hERG-channel which
made it unsuitable for further development due to the high prob-
ability of adverse side effects.20,21 With the aim of developing
new ligands of the ifenprodil binding site of the NMDA receptor,
we extended our studies to the isoindoline derivatives 8–11
(Scheme 2) exhibiting structural features similar to those of the
isoquinoline derivative 7 except that the isoquinoline ring has been
replaced by an isoindoline moiety. The methoxy groups present in
the isoquinoline derivatives had been omitted as the structural
features seemed to be sufficient to get first SAR of the isoindoline
{N-[2-(4-hydroxyphenyl)ethyl]-5-phenylpentylamine}15 and
5
[( )-3-(4-hydroxyphenyl)-1-(5-phenylpentyl)pyrrolidine].14
Furthermore, it has been shown that the potency and selectivity
of antagonists is strongly influenced by the stereochemistry of the
⇑
Corresponding author. Tel.: +49 89 2180 77249; fax: +49 89 2180 77247.
E-mail address: klaus.wanner@cup.uni-muenchen.de (K.T. Wanner).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.119

5796
A. Müller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5795–5799
OH
N
HO
Cl
N
OH
H
3C
F
erythro-Ifenprodil (1, racemate)
(
R)-Eliprodil (2)
H
O
OH
CH3
N
H
H
N
O
4
Ro 25-6981 (3)
N
O
H
OH
HO
OH
5
N
H3C
O
O
CH3
CH3
(+)-CP-101.606 (6)
N
1'S
H
3C
OH
1S
Cl
Scheme 1. Ligands of the ifenprodil binding site of the NMDA receptor (selection).
7
H
3C N
H3C N
H 3C
N
H
3C
N
OH
OH
OH
OH
R
R
R
R
8a-c
9a-c
a: R = CH3; b: R = NO2; c: R = Cl
Scheme 2. Isoindoline derivatives as new ligands for the ifenprodil binding site of the NMDA receptor.
10a-c
11a-c
scaffold. These should indicate whether the intended structural
changes are able to reduce the hERG channel activity without hav-
ing a significant effect on the potency at the ifenprodil binding site,
thus providing access to compounds with an improved pharmaco-
logical profile (Scheme 2).
In case of the isoquinoline derivatives, the stereochemistry of
the compounds had a profound influence on their affinity for the
ifenprodil binding site of the NMDA receptor. A similar behavior
was expected for the isoindoline derivatives. For this reason, we in-
tended to study the biological activity of the individual stereoiso-
mers. Accordingly, we developed an asymmetric synthesis that
provides the target compounds 8–11 in diastereomerically and
enantiomerically pure form. So far we have already published the
synthesis of the two diastereomeric series 10a–c and 11a–c in
which the configuration of the stereocenter in the isoindoline ring
is identical [(S)] but opposite in the side chain.22 The synthesis of
these compounds was based on the use of the diastereomerically
and enantiomerically pure imido ketones 12a–c that had been pre-
pared by asymmetric
a-amidoalkylation reactions (Scheme 3). It
was expected that the strategy, already successfully used for the
preparation of 10a–c and 11a–c, should be applicable to the syn-
thesis of stereoisomers 8a–c and 9a–c. For this purpose instead
of the imido ketones 12a–c the diastereomeric imido ketones
13a–c were employed as starting materials, the preparation of
which has been published earlier.22
Accordingly, in the first part of the synthesis, the imido ketones
13a–c had to be transformed into the indolinone derivatives 16a–c
that should serve as common intermediates for the preparation of
the target compounds 8a–c and 9a–c.23 As for 12a–c, the reduction
of the keto function of diastereomers 13a–c proceeded best by
treating the compounds with Li[Al(tBuO)3H] in THF at 0 °C for
16 h. Under these conditions, the reaction went to completion

A. Müller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5795–5799
5797
O
O
O
O
O
ii
v, vi
iii
N
O
O
i
O
N
H
N
N
N
HN
O
H
R*
R*
R*
H
OR'
O
OH
O
OH
R
19a-c
R
R
R
R
R
(major isomer)
17a (R' = -COPh)
18a (R' = -H)
15a-c
+
13a-c
12a-c
16a-c
vii
iv
ii
v, vi
17b-c (R' = -COPh)
18b-c (R' = -H)
O
O
R*
N
C
H3
N
H
3C
N
OH
N
H
vii
10a-c
11a-c
O
OH
OH
H
R
R
a
CH3
14a-c
R
R
R
R*
=
b
c
NO2
Cl
9a-c
8a-c
20a-c
15a/14a = 92 : 8
15b/14b = 93 : 7
15c/14c = 93 : 7
O
O
Scheme 3. Reactions and conditions: (i) Li[Al(tBuO)3H], THF, 0 °C; (ii) LiBH4, THF; (iii) DIAD, PPh3, benzoic acid, THF; (iv) Li[AlH2(OCH3)2], THF; (v) BH3ꢂS(CH3)2, THF, reflux;
(vi) HCHO, THF; (vii) NaBH3CN, CH3OH, HOAc.
and proceeded with high diastereoselectivity yielding a mixture of
diastereomers in which 15a–c predominated in a ratio of 92:8 to
93:7. The yields for the major diastereomers isolated in pure form
amounted to 79–89%. Subsequent reductive removal of the chiral
auxiliary with LiBH4 in THF provided the intermediates 16a–c
(yields 45–53%).
11a and the chloro derivative 11c, both possessing (1S,10S)-config-
uration, exhibited the highest potencies with their IC50 values
amounting to 0.43 0.06 and 0.27 0.07, respectively.
Furthermore, the functional inhibitory effects of compounds
8a–c, 9a–c, 10a–c, and 11a–c on [3H]MK-801 binding to open
NMDA receptor channels under nonequilibrium conditions were
studied. Antagonists like ifenprodil or eliprodil are known to effect
[3H]MK-801 binding in a biphasic manner. The high affinity phase
is thought to reflect the allosteric inhibition at NR2B subunit con-
taining NMDA receptors, whereas the low affinity phase is as-
sumed to account for inhibition of NMDA receptors lacking NR2B
subunits. With two exceptions, 11a and 11c, the compounds tested
inhibited [3H]MK-801 binding only with low potency. Further-
more, the same compounds gave rise to monophasic inhibition
curves, which is likely to be due to low subtype selectivity for
NR2B subunit containing NMDA receptors.
Compounds 16a–c provided direct access to target compounds
8a–c. The diastereomers 9a–c was prepared from the respective
precursor 16a–c with an additional step for the stereoinversion
of the asymmetric center in the side chain. Thus, treatment of
16a–c with borane-dimethyl sulfide (BMS) for the reduction of
the lactame moiety and subsequently with formaldehyde without
prior isolation of the formed amino alcohols that appeared difficult
to purify gave the oxazine derivatives 20a–c (yields 53–58%). Final-
ly reductive ring cleavage with NaBH3CN, provided the target com-
pounds 8a–c (yields 46–70%). To change the (S)-configuration of
the stereocenter in the side chain to (R) 16a–c were subjected to
Mitsunobu esterification (with DIAD, PPh3, benzoic acid) giving
17a–c (yields 37–77%). Subsequent reductive cleavage of the ester
function present in 17a–c provided the diastereomers 18a–c differ-
ing from 16a–c by the (R)-configuration of the stereocenter in the
side chain. Final transformations were performed as for the inter-
conversion of 16a–c into 8a–c yielding the series of target com-
pounds 9a–c. For the pharmacological tests the hydrochlorides of
these amino alcohols were used.
In contrast, inhibition of [3H]MK-801 binding by the isoindoline
derivatives 11a and 11c occurred in a biphasic manner with IC50
values for the high affinity phase close to the IC50 values observed
in the [3H]ifenprodil displacement assay (Fig. 1). These data are in
support of the subtype selectivity of 11a and 11c for NR2B subunit
containing NMDA receptors. These results are in accord with the
finding, that for antagonists selective for this receptor subtype
the potency of [3H]ifenprodil displacement correlates with the
high affinity phase of [3H]MK-801 binding.
The potency of compounds 8a–c, 9a–c, 10a–c, and 11a–c as li-
gands of the ifenprodil site of the NMDA receptor was determined
in competitive binding assays based on [3H]ifenprodil as radioli-
gand.23,24 As can be seen from the results summarized in Table 1,
the absolute configuration of the test compounds had a pro-
nounced influence on the affinity. Whereas the (1R,10S)-stereoiso-
mers 10a–c exhibited the lowest potency, which slightly
improved for the (1R,10R)- and (1S,10R)-configured isomers 9a–c
and 8a–c, the highest affinity was observed for 11a–c exhibiting
(1S,10S)-stereochemistry. Additionally, the substituent present in
p-position on the phenyl ring in the side chain affected the potency
to some extent.
According to these results, the stereochemical requirements for
binding to the ifenprodil binding site of the NMDA receptor are
preserved when the isoquinoline scaffold in 7 is replaced by an iso-
indoline moiety as in 8–11. In both series, the most potent inhibi-
tors are among the compounds possessing (1S,10S)-configuration,
that is, 7, 11a, and 11c. In addition, these compounds have distinct
subtype selectivity for NR2B subunit containing NMDA receptors
in common. But the switch from the isoquinoline skeleton to the
isoindole moiety clearly affected the potency reducing it in the
case of the p-chloro phenyl derivatives by a factor of ꢁ5 (compare
7 and 11c).
Compounds 11a and 11c were finally evaluated for their affinity
to the hERG channel (subcloned from hERG, GenBank Accession
No. U04270) expressed in Chinese hamster ovary cells using patch
clamp electrophysiology.23
Thus, within the four sets of stereoisomers the nitro substituted
compounds 8b–11b were of equal or lower potency than their
chloro and methyl counter parts. Overall, the methyl derivative

5798
A. Müller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5795–5799
Table 1
a
Results of the pharmacological tests
100
R
IC50 SEM
IC50 SEM
11a
11c
[3H]ifenprodil [3H]MK-801
[lM] (n = 3)
[l
M] (n = 3)
8aꢂHCl CH3 9.3 1.3
8bꢂHCl NO2 42 10
23
43 10
47 14
4
C N
1'R
50
0
H3
H3
H3
8cꢂHCl Cl 21
4
OH
1S
x HCl
-9
-8
-7
-6
-5
-4
-3
R
log c Inhibitor
9aꢂHCl CH3 35 19
9bꢂHCl NO2 52 10
31
5
82%a
C N
1'R
9cꢂHCl Cl 11
5
78%a
b
OH
100
50
0
1R
11a
11c
x HCl
R
10aꢂHCl CH3 65%a
10bꢂHCl NO2 54%a
28
39
26 10
4
7
C N
1'S
10cꢂHCl Cl 77%a
OH
1R
-10
-9
-8
-7
-6
-5
-4
-3
x HCl
log c Inhibitor
R
11aꢂHCl CH3 0.43 0.06
11bꢂHCl NO2 7.5 0.3
0.21 0.05b 20 4c
Figure 1. Inhibition of [3H]ifenprodil binding (a) and [3H]MK-801 binding (b) in
presence of 100 -glutamate and 30 M glycine to a synaptosomal fraction of
H3C N
1'S
43 15
lM
L
l
11cꢂHCl Cl 0.27 0.07
0.50 0.12b 67 14c
porcine hippocampus, by 11a and 11c, respectively. Data points represent
means SD (calculated from triplicates) from one representative experiment out
of three.
OH
1S
x HCl
similar manner to the related isoquinoline derivatives, for exam-
ple, 7, by the ifenprodil binding site. Though, the isoindoline deriv-
atives do not reach the potency of their isoquinoline counterparts,
for the two most active compounds 11a and 11c the potency is
only about 5–10 times lower than that of 7. In common with iso-
quinolines like 7 the isoindoline derivatives 11a and 11c display
distinct subtype selectivity for NR2B subunit containing NMDA
receptors but, unfortunately, also a reasonable potency as inhibi-
tors of the hERG channel.
R
O CH3
O CH3
N
1'S
H3C
OH
0.059 0.0202
717
0.103 0.04b 23.5 3.4c
1S
22.5 2.2c
x HCl
Cl
Specific binding at 100
a
l
M in comparison to a control experiment without
Supplementary data
inhibitor.
b
High affinity fraction.
Low affinity fraction.
c
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.07.119.
Unfortunately, both compounds strongly blocked hERG cur-
rents, to 2.6 2.4% (11a, 100 M) and 1.6 2.0% (11c, 100 M) of
References and notes
l
l
1. (a) Muir, K. W.; Lees, K. R. Stroke 1995, 26, 503; (b) Rogawski, M. A. Drugs 1992,
34, 106.
2. Utech, T.; Köhler, J.; Wünsch, B. Eur. J. Med. Chem. 2011, 46, 2157.
3. Bräuner-Osborne, H.; Egebjerg, J.; Nielsen, E. O.; Madsen, U.; Krogsgaard-
Larsen, P. J. Med. Chem. 2000, 43, 2609.
4. Stawski, P.; Janovjak, H.; Trauner, D. Bioorg. Med. Chem. 2010, 18, 7759.
5. Scatton, B.; Giroux, C.; Thenot, J. P.; Frost, J.; George, P.; Carter, C.; Benavides, J.
Drugs Future 1994, 19, 905.
6. Mony, L.; Triballeau, N.; Paoletti, P.; Acher, F. C.; Bertrand, H.-O. Bioorg. Med.
Chem. Lett. 2010, 20, 5552.
control, which in turn would be predicted to cause an increase in
the Q/T interval in the ECG.25 As such an effect could be dangerous
and lead to sudden cardiac arrest, especially in subgroups of pa-
tients with a genetic disposition to oversensitivity or under condi-
tions of altered extracellular K+ concentration, similar to 7, both
most active compounds, 11a and 11c, appear to be unsuitable for
use as pharmaceuticals.21
In summary, four sets of stereoisomeric isoindoline derivatives,
8a–c to 11a–c, have been synthesized and evaluated for their po-
tency in [3H]ifenprodil displacement assays and functional inhibi-
tion of [3H]MK-801 binding under nonequilibrium conditions. As
for related isoquinoline derivatives represented by the prototypic
compound 7 the potency was strongly dependent on the absolute
configuration. The highest potency was observed for isoindoline
derivatives displaying, like 7, (1S,10S)-stereochemistry. The results
implicate that the isoindoline derivatives are recognized in a
7. Nguyen, K. T.; Claibrone, C. F.; McCauley, J. A.; Libby, B. E.; Claremon, D. A.;
Bednar, R. A.; Mosser, S. D.; Gaul, S. L.; Connolly, T. M.; Condra, C. L.; Bednar, B.;
Stump, G. L.; Lynch, J. L.; Koblan, K. S.; Liverton, N. J. Bioorg. Med. Chem. Lett.
2007, 17, 3997.
8. Alarcon, K.; Martz, A.; Mony, L.; Neyton, J.; Paoletti, P.; Goeldner, M.; Foucaud,
B. Bioorg. Med. Chem. Lett. 2008, 18, 2765.
9. Tewes, B.; Frehland, B.; Schepmann, D.; Schmidtke, K.-U.; Winckler, T.;
Wünsch, B. Bioorg. Med. Chem. 2010, 18, 8005.
10. Williams, K. Mol. Pharmacol. 1993, 44, 851.
11. Perin-Dureau, F.; Rachline, J.; Neyton, J.; Paoletti, P. J. Neurosci. 2002, 22, 5955.
12. Toulmond, S.; Serrano, A.; Benavides, J.; Scatton, B. Brain Res. 1993, 620, 32.

A. Müller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5795–5799
5799
13. Fischer, G.; Mutel, V.; Trube, G.; Malherbe, P.; Kew, J. N. C.; Mohacsi, E.; Heitz,
M. P.; Kemp, J. A. J. Pharmacol. Exp. Ther. 1997, 283, 1285.
Nielsen, J.; Pagnozzi, M. J.; Richter, K. E. G.; Ronau, R. T.; Shalaby, I. A.; Stemple,
J. Z. J. Med. Chem. 1995, 38, 3138.
14. Guzikowski, A. P.; Tamiz, A. P.; Acosta-Burruel, M.; Hong-Bae, S.; Cai, S. X.; Ha, J.
E.; Keana, J. F. W.; Kesten, S. R.; Shipp, C. T.; Tran, M.; Whittemore, E. R.;
Woodward, R. M.; Wright, J. L.; Zhou, Z.-L. J. Med. Chem. 2000, 43, 984.
15. Tamiz, A. P.; Whittemore, E. R.; Zhou, Z.-L.; Huang, J.-C.; Drewe, J. A.; Chen, J.-C.;
Cai, S.-X.; Weber, E.; Woodward, R. M.; Keana, J. F. W. J. Med. Chem. 1998, 41, 3499.
16. Chenard, B. L.; Shalaby, I. A.; Koe, B. K.; Ronau, R. T.; Butler, T. W.; Prochniak, M.
A.; Schmidt, A. W.; Fox, C. B. J. Med. Chem. 1991, 34, 3085.
18. Pabel, J.; Höfner, G.; Wanner, K. T. Bioorg. Med. Chem. Lett. 2000, 10, 1377.
19. Wanner, K. T.; Praschak, I.; Höfner, G.; Beer, H. Arch. Pharm. 1996, 329, 11.
20. Höfner, G.; Hoesl, C. E.; Parsons, C.; Quack, G.; Wanner, K. T. Bioorg. Med. Chem.
Lett. 2005, 15, 2231.
21. Sanguinetti, M. C.; Tristani-Firouzi, M. Nature 2006, 440, 463.
22. Müller, A.; Polborn, K.; Wanner, K. T. J. Heterocycl. Chem. 2007, 44, 575.
23. See Supplementary data.
17. Chenard, B. L.; Bordner, J.; Butler, T. W.; Chambers, L. K.; Collins, M. A.; De
Costa, D. L.; Ducat, M. F.; Dumont, M. L.; Fox, C. B.; Mena, E. E.; Menniti, F. S.;
24. Höfner, G.; Wanner, K. T. Eur. J. Pharmacol. 2000, 394, 211.
25. Smith, P. L.; Baukrowitz, T.; Yellen, G. Nature 1996, 379, 833.