2-Styryl-pyridines and 2-(3,4-Dihydro-naphthalen-2-yl)-pyridines as potent NR1/2B subtype selective NMDA receptor antagonists

Article abstract of DOI:10.2533/000942904777677579

A series of 2-styryl-pyridines and 2-(3,4-dihydro-naphthalen-2-yl)- pyridines was prepared and evaluated as NR1/2B subtype selective NMDA receptor antagonists. The SAR developed in this series resulted in the discovery of high affinity antagonists that are selective (vs. α₁ and M ₁ receptors) and are active in vivo.

Full text of DOI:10.2533/000942904777677579

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
630
CHIMIA 2004, 58, No. 9
Chimia 58 (2004) 630–633
© Schweizerische Chemische Gesellschaft
ISSN 0009–4293
2-Styryl-pyridines and
2-(3,4-Dihydro-naphthalen-2-yl)-
pyridines as Potent NR1/2B Subtype
Selective NMDA Receptor Antagonists
Bernd Buettelmanna*, Alexander Alaninea, Anne Boursonb, Ramanjit Gillb, Marie-Paule Heitza,
Vincent Mutelb, Emmanuel Pinarda, Gerhard Trubeb, and René Wylera
Abstract: A series of 2-styryl-pyridines and 2-(3,4-dihydro-naphthalen-2-yl)-pyridines was prepared and evaluat-
ed as NR1/2B subtype selective NMDA receptor antagonists. The SAR developed in this series resulted in the
discovery of high affinity antagonists that are selective (vs. α1 and M1 receptors) and are active in vivo.
Keywords: Neurodegeneration · NMDA · NR1/2B antagonists · Structure-activity relationship
Introduction
As a result of a medicinal chemistry pro- trans-styryl and the 1,2-dihydronaphtalene
gram to discover structurally novel NR1/2B motif (as in 3, 4) turned out to be the most
Whilst NMDA receptor-mediated gluta- subtype selective NMDA antagonists, we promising surrogates. In this communica-
matergic neurotransmission is essential for recently disclosed [7] tetrahydroisoquino- tion we would like to report on our efforts to
normal brain function, its overexcitation is line (THI) substituted pyridines (e.g. 2) and evaluate these novel structural series.
implicated with the pathogenesis of neu- THI substituted quinolines (Table 1). Since
rodegenerative diseases such as stroke and there are some toxicological concerns about
Parkinson disease as well as depression [1] the THI substructure [8], early on we sought Chemistry
(Table 1). Over the past decade evidence a bioisosteric replacement for this moiety.
has accumulated that antagonists of the Among the different isosteric structures
The unsubstituted trans-2-styryl-pyri-
NR1/2B subtype of the NMDA receptor suggested by molecular modeling, the dine 5 as well as its 6-amino substituted
can be expected to combine efficacy in the
treatment of these diseases with an overall
Table 1. Binding affinities of reference compounds
acceptable side-effect profile [4]. In 1993
the antihypertensive Ifenprodil (1) was
identified as the first NR1/2B subtype se-
lective antagonist [5]. As the α1 mediated
cerebrovascular effect of 1 is contraindicat-
ed in stroke, much effort has been devoted
towards identifying structurally related
compounds with an increased selectivity
NMDA vs the adrenergic α1 receptors [6].
*Correspondence: Dr. B. Buettelmanna
Tel.: +41 61 688 13 51
Fax: +41 61 688 87 14
E-Mail: bernd.buettelmann@roche.com
aPharma Division
Discovery Chemistry
F. Hoffmann-La Roche Ltd.
CH–4070 Basel
bPharma Division
Preclinical CNS Research
F. Hoffmann-La Roche Ltd.
CH–4070 Basel

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
631
CHIMIA 2004, 58, No. 9
analogue 6 are known compounds [9]. 4-
Amino substituted trans-2-styryl-pyridine
3 was synthesized by a Stille-type coupling
starting from 4-amino-2-bromopyridine
(Scheme 1). Interestingly, the 5-amino
analogue 7 could be obtained under
similar conditions starting from 2-chloro-
5-nitropyridine using 2 equiv. of trib-
utyl(phenylethenyl)tin. Under the reaction
conditions the nitro group was directly re-
duced, presumably by a tin species. The
free 6-amino function in 6 could readily be
acylated to 8 or alkylated to 9, 10 using
standard procedures. The disubstituted
trans-styryl-pyridine 11 was prepared by
close analogy to 6: According to the proto-
col developed by Honma et al. [9b][10]
trans-4-methyl-2-styryl-pyridine-1-oxide
was treated with dimethylsulfate followed
by sodium cyanide; hydrolysis of the nitrile
furnished the corresponding acid which
was subjected to a Curtius degradation
to yield trans-6-amino-4-methyl-2-styryl-
pyridine 11. The same sequence was em-
ployed for introduction of substituents in
the phenyl ring of 6. Condensation of ap-
propriately substituted benzaldehydes with
2-picoline-1-oxide gave the respective
trans-2-styryl-pyridine-1-oxides
which
were further elaborated as described above
to 12–18. Finally, ether cleavage of 4’-
methoxysubstituted 15 with boron tribro-
mide yielded the free phenol 19. In our
hands most of these styryl-pyridines had a
tendency to undergo [2+2] cycloadditions
as well as (albeit to a lesser extent) cis/trans
isomerism when exposed to sunlight. These
photochemical reactions are typically well
known [11] for stilbenes and azastilbenes.
It has been reported [12] that upon rigidifi-
cation of the styryl motif of stilbene within
a 1,2 dihydro-naphtalene ring the resulting
3-phenyl-1,2-dihydronaphtalene is photo-
chemically stable. Intrigued by this obser-
vation we sought to improve photostability
by incorporation of the 1,2 dihydro-naphta-
lene ring. Key intermediates for the envis-
aged products 4 and 20–35 were 3,4-dihy-
dro-naphthalene-2-boronic acids (Scheme
2). They could readily be prepared starting
from substituted α-tetralones by bromina-
tion, reduction and elimination to yield
the respective 3-bromo-1,2-dihydro-naph-
thalenes [13]. Halogen-metal exchange fol-
lowed by reaction with triisopropyl borate
and hydrolysis finally furnished the desired
boronic acids. The substitution pattern in
the final products was determined by the
commercial availability of the respective α-
tetralones. Under Suzuki conditions these
could be coupled to optionally substituted
2-bromopyridines [14]. As expected,
2-(3,4-dihydro-naphthalen-2-yl)-pyridines
turned out to be photochemically much
more stable when compared to trans-styryl-
pyridines. We will report elsewhere on this
aspect of reactivity in more detail.
see Table 2
Scheme 1. a) Pd(OAc)2, PPh3, DMF, 130 °C, 23–40%; b) ClCO2CH3, NEt3,
THF, rt, 83%; c) NaBH3CN, CH2O, CH3OH, rt, 22%; d) NaBH3CN, excess
CH2O, CH3OH, rt, 35%; e) 1: (subst.) benzaldehyde, KOtBu, tBuOH, 85
°C; 2: (CH3)2SO4, dioxane/THF 2:1, 50 °C; 3: NaCN, H2O, rt; 4: 25% HCl,
100 °C; 5: diphenylphosphoryl azide, NEt3, tBuOH, 85 °C; 6: 12% HCl,
EtOH, 80 °C, 1–10% (all steps); f) BBr3, CH2Cl2, rt, 25%.
Results and Discussion
plore steric bulk tolerance in this part of the
molecule, the 6-amino function in 6 was re-
Whilst the structurally most simple placed by a methylamino- and by a di-
trans-2-styryl-pyridine 5 displays only low methylamino group (9, 10). Whilst this
affinity (Ki = 1,100 nM) towards the leads to compounds with marginally re-
NR1/2B subtype of the NMDA receptor duced basicity (pKa is lowered by 0.1 and
(Table 2), introduction of electron donating 0.3 units, respectively), affinity to the NM-
substituents on the pyridine core can signif- DA receptor is markedly reduced (10 fold
icantly enhance affinity. For the amino-sub- and 200 fold, respectively), suggesting lim-
stituted 2-styryl-pyridines 3, 6, 7, and 11 in- ited size tolerance in position 6. In this re-
creasing basicity of the heterocycle paral- spect trans-2-styryl-pyridines differ from
lels an increase in binding affinity. High structurally related 2-(3,4-dihydro-1H-iso-
affinity (i.e. Ki < 20 nM) compounds 3 and quinolin-2yl)-pyridines [7a]. To address the
11 are (at least partly) protonated at physi- influence on NMDA affinity, substituents at
ological pH 7.4, suggesting that this is the the phenyl ring in 6 were varied systemati-
bioactive form interacting with the NMDA cally (12–19). No favorable lipophilic,
receptor. Replacing the 6-amino function in electronic or H-bonding interaction could
6 by a sterically more demanding methy- be identified. As even fluorine substituents
loxamate group (8) not only reduces in position 3’ and 4’ (17, 18) lead to a de-
markedly basicity (pKa = 3.1) but also NM- crease in NMDA affinity (1.5 fold and 2
DA binding affinity (Ki >38 µM). To ex- fold, respectively), it was concluded that

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
632
CHIMIA 2004, 58, No. 9
aminopyridine core with the aim of main-
taining high affinity to the NMDA receptor
whilst reducing α1 affinity. Keeping the 4-
NH2 function constant, it was found that an
additional 3-methyl group (20) not only
widens the dihedral angle about the C–C
bond connecting both cyclic systems, but
also reduces markedly NMDA affinity (Ki =
270 nM). Additional alkyl groups in posi-
tion 6 (23, 25, 26) lead to reduced NMDA
affinity combined with an increased α1
affinity, rendering these compounds less se-
lective. An additional 5-methyl group (28)
or a 6-hydroxymethyl function (4) thus
proved to be preferred substituents both in
terms of NMDA affinity and in selectivity
vs. α1. In related series we have also noted
that introduction of polar functionalities in
the side chain can lead to an increased se-
lectivity vs. α1 receptors [7]. N-methylation
of the 4-amino group (24) again leads to a
3-fold reduced NMDA binding affinity,
rendering this compound less interesting.
In the structurally related 2-(3,4-dihy-
dro-1H-isoquinolin-2yl)-pyridine series we
have identified additional muscarinic M1
receptor affinity which may lead to un-
wanted side effects [7a]. In the current 2-
(3,4-dihydro-naphthalen-2-yl)-pyridine se-
ries this potential selectivity problem
turned out to be irrelevant as all compounds
with high NMDA affinity (i.e. 4, 25, 26, 27,
28) exert M1 affinity in the micromolar
range. Having shown that polar functional-
R3 = H, R5 = see Table 2 and Table 4, resp.
Scheme 2. g) 1: Br2, ether, 10 °C; 2: NaBH4, C2H5OH/THF 1:1, rt; 3:
pTsOH, toluene, 120 °C, 32–66% (all steps); h) 1: t-BuLi, ether, –65 °C;
2: B(OiPr)3, rt; 3: 3N HCl, rt, 38–78% (all steps); i) Pd(PPh3)4, aq. K2CO3,
toluene, 100 °C, 22–65%.
the phenyl ring should optimally be unsub- supra), introduction of an additional elec-
stituted.
tron donating 4-methyl group (22) leads to
Based on these results in the trans-2- an affinity increase (Ki = 45 nM). As in the
styryl-pyridine series a similar chemical trans-2-styryl-pyridine series introduction
program was initiated to identify structural of a 4-amino-group (27) leads to a com-
determinants for NMDA affinity in the pound with high affinity to the NMDA re-
closely related 2-(3,4-dihydro-naphthalen- ceptor (Ki = 7 nM). However, 27 exhibits al-
2-yl)-pyridine series (Table 3). 6-Amino so marked affinity towards the α1 receptor
substituted 21 exerts moderate NMDA (Ki = 480 nM). Our efforts were therefore
affinity (Ki = 100 nM). As expected (vide directed towards modification of the 4-
Table 2. Binding affinities of compounds 3, 5-19. Influence of substituents at
Table 3. Binding affinities of compounds 4, 20-29. Influence of substi-
pyridine and phenyl ring.
tuents at pyridine ring.

PHARMACEUTICAL INDUSTRY IN SWITZERLAND
633
CHIMIA 2004, 58, No. 9
Table 4. Binding affinities of compounds 30-35. Influence of substituents
at dihydronaphtalene-ring.
Pulm. Renal Invest. Drugs 1999, 1(5),
576; b) J.L. Wright, T.F. Gregory, P.A.
Boxer, L.T. Meltzer, K.A. Serpa, L.D.
Wise, Bioorg. Med. Chem. Lett. 1999, 9,
2815; c) P. Skolnick, Eur. J. Pharmacol.
1999, 375, 31.
[2] V. Mutel, D. Buchy, A. Klingelschmidt, J.
Messer, Z. Bleuel, J.A. Kemp, J. Neu-
rochem. 1998, 70, 2147.
[3] P. Greengrass, R. Bremner, Eur. J. Phar-
macol. 1979, 55, 323.
[4] a) G. Fischer,A. Bourson, J.A. Kemp, H.P.
Lorez, Neuroscience Annual Meeting,
Nov. 16-21, 1996, Washington D.C.; b)
P.L. Chazot, Curr. Opin. Invest. Drugs
2000, 1, 370.
[5] K. Williams, Mol. Pharmacol. 1993, 44,
851.
[6] B.L. Chenard, F.S. Menniti, Curr. Pharm.
Des. 1999, 5(5), 381.
[7] a) B. Buettelmann, A. Alanine, A. Bour-
son, R. Gill, M.P. Heitz, V. Mutel, E.
Pinard, G. Trube, R. Wyler, Bioorg. Med.
Chem. Lett. 2003, 13, 829; b) B. Buettel-
mann, A. Alanine, A. Bourson, R. Gill,
M.P. Heitz, V. Mutel, E. Pinard, G. Trube,
R.Wyler, Bioorg. Med. Chem. Lett. 2003,
13, 1759.
Table 5. In vivo potency of selected compounds
Cpd.
ED50 [mg/kg]a
Sound induceded seizures
[8] T. Niwa, M. Kajita, T. Nagatsu, Pharma-
col. Endog. Neurotoxins 1998, 3.
1
2
16
13
7
[9] a) J.L.R. Williams, S.K. Webster, J.A. Van
Allan, J. Org. Chem. 1961, 26, 4893; b)Y.
Honma, K. Hanamoto, T. Hashiyama, Y.
Sekine, M. Takeda, O. Yasutoshi, K.
Tsuzurahara, J. Med. Chem. 1984, 27,
125.
[10] A. Silhankova, J. Hulvova, P. Trska, M.
Ferles, Collect. Czech. Chem. Commun.
1989, 54, 1687.
4
6
27
15
27
a Compounds were administered i.p. 30 min.
before testing in DBA/2 mice.
[11] a) J.Vansant, S. Toppet, G.J. Smets, J. Org.
Chem. 1980, 45, 1565; b) G. Marconi, G.
Bartocci, U. Mazzucato, A. Spalletti, F.
Abbate, L. Angeloni, E. Castellucci,
Chem. Phys. 1995, 196, 383.
[12] J.J.M. Lamberts, W.H. Laarhoven, Recl.
Trav. Chim. Pays-Bas 1984, 103, 131.
[13] M. Adamczyk, D.S. Watt, D.A. Netzel, J.
Org. Chem. 1984, 49, 4226.
[14] For the synthesis of substituted 2-bro-
mopyridines, see [5].
[15] A. Avdeef, ‘Applications and Theory
Guide to pH-Metric pKa and logP Mea-
surement’, Sirius Analytical Instruments
Lts.: Forest Row UK, 1993.
[16] R.L. Hudkins, J.F. Stubbins, D.L. De-
Haven-Hudkins, Eur. J. Pharmacol. 1993,
231, 485.
[17] A. Bourson, V. Kapps, C. Zwingelstein, A.
Rudler, F.G. Boess, A.J. Sleight, Naunyn-
Schmiedeberg’s Arch. Pharmacol. 1997,
356, 820.
ities in position 6 (CH2OH in 4, NH2 in 22)
are tolerated at the NMDA receptor, pyri-
done 29 served as a test of our working hy-
pothesis, namely that pyridines in the pro-
tonated form act as the bioactive substruc-
ture. Under physiological conditions 29 is
uncharged but projects an amid proton for
potential interaction. As 29 has no NMDA
affinity, we conclude that the positive
charge at the heterocycle is the primary
structural determinant for binding affinity.
Finally substituent effects at the 3,4-di-
hydro-naphthalene were studied (Table 4).
Keeping the 4-amino function in 27 con-
stant, introduction of a 7’-Cl or 4’-CH3 (34,
35) leads to compounds with high NMDA
affinity, however, compared to 27, selectivi-
ty vs. α1 is significantly reduced. Other sub-
stituents explored (30–33) proved to be less
interesting with regard to NMDA affinity.
In vivo activity of key compounds was
measured in mice after i.p. administration
using the standard sound-induced seizures
model [17]. As depicted in Table 5, trans-
styryl-pyridine 6 was somewhat less active
when compared to the references 1 and 2.
On the other hand, dihydro-naphtalene sub-
stituted pyridines 4 and 27 were at least as
potent as the references, indicating an ade-
quate brain exposure of these compounds.
Conclusion
Starting from the recently described
NR1/2B subtype selective NMDA antago-
nist 2 we have characterized a series of 2-
styryl-pyridines and 2-(3,4-dihydro-naph-
thalen-2-yl)-pyridines that follow a similar
SAR. This led to the identification of 4, a
compound that is active in vivo, combining
high affinity at the NMDA receptor with
low muscarinic and adrenergic side effect
liabilities.
Received: June 18, 2004
[1] a) R. Gill, J.A. Kemp, J.G. Richards,
J.N.C. Kew, Curr. Opin. Cardiovasc.,