Arylmethoxypyridines as novel, potent and orally active mGlu5 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.12.088

Optimisation of affinity, chemical stability, metabolic stability and solubility led from a chemically labile HTS hit 1 to mGlu5 receptor antagonists (24-26) with high affinity for the allosteric MPEP binding site, improved microsomal metabolic stability and anxiolytic-like activity in vivo as assessed by the Vogel conflict drinking test.

Full text of DOI:10.1016/j.bmcl.2005.12.088

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1892–1897
Arylmethoxypyridines as novel, potent and orally active
mGlu5 receptor antagonists
a,
a
a
Bernd Buttelmann,a,* Jens-Uwe Peters, Simona Ceccarelli, Sabine Kolczewski,
*
¨
Eric Vieira,a Eric P. Prinssen,b Will Spooren,b Franz Schuler,b Jo¨rg Huwyler,b
Richard H. P. Porterb and Georg Jaeschkea
aPharma Division, Discovery Chemistry, F. Hoffman-La Roche Ltd, CH-4070 Basel, Switzerland
bPharma Division, Preclinical CNS Research, F. Hoffman-La Roche Ltd, CH-4070 Basel, Switzerland
Received 16 November 2005; revised 27 December 2005; accepted 28 December 2005
Available online 24 January 2006
Abstract—Optimisation of affinity, chemical stability, metabolic stability and solubility led from a chemically labile HTS hit 1 to
mGlu5 receptor antagonists (24–26) with high affinity for the allosteric MPEP binding site, improved microsomal metabolic stability
and anxiolytic-like activity in vivo as assessed by the Vogel conflict drinking test.
Ó 2006 Elsevier Ltd. All rights reserved.
L-Glutamate, the major excitatory neurotransmitter in
the CNS, exerts its effects by stimulation of two major
types of glutamate receptors, the ionotropic glutamate
receptors (iGluRs) and the metabotropic glutamate
receptors (mGlu receptors).1 Whereas the functional
role of the iGluRs has been thoroughly investigated,
the mGlu receptors have attracted increasing interest
only in recent years. mGlu receptors are class C GPCRs,
characterised by a large N-terminus incorporating the
ligand binding domain and are divided into eight sub-
types, mGlu1- mGlu8.2 Among the mGlu receptors,
the mGlu5 receptor is expressed in the limbic areas of
the brain, which suggests a potential role of this receptor
in psychiatric disorders, such as anxiety.3 Indeed, the
selective mGlu5 receptor antagonists MPEP and MTEP
have been shown to be active in a broad range of
preclinical anxiety tests.4 Recently, the clinically validat-
ed anxiolytic Fenobam was found to be a selective,
high-affinity mGlu5 receptor ligand. MPEP, MTEP
and Fenobam bind to a unique allosteric site in the
transmembrane region of mGlu5 receptors, which
allows for a high selectivity of these compounds espe-
cially over other glutamate receptors.5 The limitations
of current anxiety treatments, such as side effects
(benzodiazepines) or slow onset of action (SSRIs), have
provided the impetus for the discovery of novel, and
potentially superior, anxiolytics.6 In this regard, the
mGlu5 receptor appears attractive as an anxiety target,
because it offers good chemical tractability, the potential
for high selectivity via a unique allosteric site and a val-
idated proof of concept.7 Additionally, mGlu5 receptor
antagonism has been proposed as a potential target for
the treatment of pain, obesity, Parkinson’s disease and
drug abuse.8 Consequently, the mGlu5 receptor has
attracted much attention, and several analogues of
MPEP/MTEP have been published as mGlu5 receptor
antagonists.9 We report here our investigation of a novel
series of non-MPEP-like mGlu5 receptor antagonists.
We identified the phthalimide of a benzyloxyphenyl-
amine 1 from the Roche compound collection in a
high-throughput screen, which was based on the dis-
placement of [3H]MPEP from recombinantly expressed
mGlu5 receptors. Some first exploratory chemistry
around 1 revealed that the methyl substituent R
(Fig. 1) is essential, and analogues without a substituent
in this position, such as 2, have a greatly decreased affin-
ity. An improvement of in vitro affinity (for measurement
details, see 10) was achieved by introduction of a chloro
or cyano substituent into the meta-position of the benzyl-
ic ring (Table 1) leading to compounds 3 and 4, whereas
substituents in the para position were not tolerated.
However, a main disadvantage of compounds 1–8 (Table
1) is their susceptibility towards hydrolysis of the phthal-
imide motif over a wide pH range.
Keywords: mGluR5; Metabotropic glutamate receptor; Anxiolytic;
Anxiolysis.
*
Corresponding authors; e-mail addresses: bernd.buettelmann@
roche.com; jens-uwe.peters@roche.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.12.088

B. Bu¨ttelmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1892–1897
1893
Table 2. Exploration of substituents R to replace the chemically
susceptible phthalimide motif (1–8)
N
N
N
S
MPEP
Ki = 4nM
MTEP
Ki = 6nM
Cl
O
R
R
O
H
H
R=
Ki (nM)
Cl
N
N
N
O
N
N
O
N
O
9
10
11
12
13
14
160
O
(HTS hit)
Ki = 51nM
Ki = 2100nM
1: R = CH3
2: R = H
Fenobam
Ki = 61nM
O
O
O
nC5H11
33
48
70
27
58
N
N
O
F
Figure 1. Ki values (hmGlu5 receptor, as determined in-house, see
Ref. 10) for the prototypical non-competitive mGluR5 antagonist
MPEP, its analogue MTEP, the anxiolytic Fenobam, our HTS hit
1 and its des-Me analogue 2.
F
O
N
N
Table 1. Exploration of substituents R
N
N
O
O
N
R
O
R=
Ki (nM)
3
4
5
6
7
8
m-Cl
m-CꢀN
p-Cl
4.7
6.0
Table 3. Introduction of nitrogen into different positions of the central
ring
3900
>10,000
1500
900
p-CH3
p-MeO
p-F
X
Y
Cl
O
Z
N
Compounds 3 and 4 have a 10-fold improved affinity as compared to 1.
X
Y
Z
Ki (nM)
Therefore, numerous replacements of the phthalimide
moiety were examined, out of which 9–14 (Table 2)
emerged as being best in terms of mGluR5 affinity.
These compounds however had unfavourable molecular
properties, such as high lipophilicity, low stability in rat
liver microsome preparations11 and low solubility
9
15
16
17
CH
N
CH
CH
N
CH
CH
CH
N
160
45
CH
CH
>3000
>3000
CH
(logD > 3.0, CLint > 100 ll/min/lg protein, LYSA12
1 lg/ml).
<
of 1 nM. At this point, some structural similarities be-
tween the arylmethoxypyridines and published MPEP
analogues became apparent: in both series, an aryl ring
is linked by a 2-carbon spacer to an a-pyridine. Further-
more, small, lipophilic meta substituents on the aryl ring
increased the affinity in both series, although this effect is
more pronounced for the arylmethoxypyridines.13 On
the other hand, the different geometry of the methoxy
linker and the acetylene linker is expected to lead to dif-
ferent spatial orientations of the aryl and the pyridine
moieties for the two series. The most obvious SAR dif-
ferences between the two series include the pyridyl-nitro-
gen in the 2-position, which seems to be mandatory for
activity in the MPEP series but not for the arylmethoxy-
pyridines, and the substituent in the a0-position of this
nitrogen, which is stringently required for high affinity
in the arylmethoxypyridine-, but not in the MPEP,
series.14
It was anticipated that compounds with decreased lipo-
philicity might have improved microsomal stability and
improved solubility. Following this hypothesis, the cen-
tral aromatic motif was replaced by all possible pyri-
dines. Pyrrole as an isostere for phthalimide was
selected as model system for these optimisation studies
due to its straightforward synthetic accessibility.
The introduction of nitrogen into the Y (16) and Z (17)
positions was not tolerated (Table 3). However, a nitro-
gen in the X position (15) did not only decrease the cal-
culated lipophilicity (clogP = 6.2 [15] vs 7.1 [9]), but
gave a further improvement in affinity. The introduction
of the nitrogen in the X position had changed the
requirements for the R substituent (18–23, Table 4).
Among the best substituents in this position were now
a trifluoroacetamide 18 (Table 4), a ‘reversed’ amide
19 and especially a nitrile 20 with an outstanding affinity
A single-dose pharmacokinetic study revealed that 20
achieved only low plasma levels after oral application,

1894
B. Bu¨ttelmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1892–1897
Table 4. Exploration of substituents R to further improve the mGluR5
affinity
Similar to the considerations outlined earlier, we hoped
to further reduce logD, and thereby increase solubility,
by replacing the lipophilic chlorobenzyloxy motif by
an appropriately substituted picolinyl motif. Additional-
ly, the electron density of the pyridine nitrogen would be
lowered further, rendering metabolism at this site more
unlikely. This was realized with 25, which retained most
of the affinity of its predecessor 24, and was soluble,
albeit to a limited extent. A further improvement in sol-
ubility was made with 26, in which the basicity of the
peripheral pyridine moiety was increased by replacing
the meta-chloro with a meta-methyl substituent. Com-
pounds 20 and 25 behaved as functional antagonists in
a FLIPR assay (IC50 = 9 nM; 72 nM, respectively).
Compounds 20 and 24–26 were evaluated in the Vogel
conflict test, in which drinking is suppressed in water-re-
stricted rats by a brief electrical shock every second of
cumulative drinking time.15 Both 20 and 26, after an
oral dose of 10 mg/kg, significantly increased drinking
time, consistent with an anxiolytic-like profile (Fig. 2).
Compounds 24 and 25 showed similar effects after an
oral dose of 15 and 10 mg/kg, respectively (data not
shown). It is assumed that the somewhat reduced affin-
ities of 24–26 are counterbalanced by improved PK
parameters (in comparison with 20).
N
Cl
O
R
R=
Ki (nM)
O
18
19
20
21
22
18
CF3
N
N
27
1
N
N
O
N
11
28
N
N
OH
N
O
N
N
23
44
***
NH2
100
O
80
**
***
**
due to its intermediate-to-high clearance and its high
volume of distribution (Table 5). Furthermore, the low
solubility of 20 might contribute to a poor absorption
from the GI tract. An HPLC/MS analysis of the metab-
olites obtained from an incubation of 20 with rat liver
microsomes suggested that the pyridine nitrogen and
possibly the adjacent methyl group are the main sites
of oxidative metabolism. Interestingly, and contrary to
our expectations, the benzylether methylene seemed
not to be significantly involved in microsomal metabo-
lism. We anticipated that the replacement of the pico-
line-methyl with a chloro substituent might improve
the metabolic stability by removing the methyl substitu-
ent as a metabolic weak site, and by reducing the elec-
tron density in the adjacent nitrogen. Indeed, 24
showed a reduced intrinsic clearance with rat micro-
somes. However, 24 still had a poor solubility which
was considered detrimental to good oral bioavailability.
60
40
20
0
vehicle MPEP
vehicle MPEP
20
26
(
)
(
)
(
)
(
)
26
20
20
26
Figure 2. Drinking times (s) in the Vogel conflict test15 (punished
drinking) as a measure of anxiolytic-like response. The drinking times
after an oral dose of 10 mg/kg of 20 and 26 are compared with vehicle
and MPEP (10 mg/kg po) as positive control. Data are median and
interquartile range; Statistics: **p < 0.01; ***p < 0.001 vs vehicle
(Mann–Whitney U test, one-tailed).
O
a, b
Cl
O
N
HO
NH2
Table 5. Pharmacokinetic properties of 20
O
27
3
9
N
c, d
27
Cl
O
N
Cl
O
N
20
CL (rat) = 43 ml/min/kg
Vss = 14 l/kg
t1/2 (iv) = 5 h
F = 4.7%
Cmax (10 mg/kg po) = 86 ng/ml
Brain/plasma = 3.6
Scheme 1. Reagents and conditions: (a) phthalic anhydride, acetic
acid, 1.5 h reflux; (b) 3-chlorobenzyl chloride, Cs2CO3, DMF, 1.5 h
100 °C; (c) NaH, DMF, 45 min rt, then 3-chlorobenzyl bromide, 12 h
rt; (d) acetonylacetone, pTsOH Æ H2O, toluene, 1.5 h reflux (water
trap).
Arrows indicate the putative sites of oxidative metabolism as suggested
by a microsomal metabolite analysis.

B. Bu¨ttelmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1892–1897
1895
Compounds 3–14 (Tables 1 and 2) were prepared from
3-amino-5-methyl-phenol (27).16 For instance, 3 was pre-
pared by phthalimide formation and subsequent benzyla-
tion, whereas 9 was prepared by benzylation in a first step,
followed by a Paal–Knorr reaction (Scheme 1). Pyrroles
15–17 (Table 3) were obtained by the synthetic sequences
outlined in Scheme 2: a 4-pyrrole substituent was intro-
duced onto 2-bromo-5-methylpyridinium-N-oxide (28)17
by a nitration/reduction sequence (29, 30) followed by a
Paal–Knorr reaction to give 31. The isomeric building
block 35 was available from epoxide 32 via 33 and
34. Both building blocks 31 and 35 were reacted with 3-
chlorobenzyl alcohol to afford 15 and 17, respectively.
Compound 16 was obtained by first introducing the
benzyloxy substituent into a picoline N-oxide 36 to give
37, and subsequent introduction of the pyrrol motif via
38 and 39.
O
N+
O
N+
a
b
Br
NO2
Br
28
29
N
c
N
Br
N
Br
NH2
30
31
Nitriles 20 and 24 (Table 4) were synthesized from isoni-
cotinic acid derivatives 40 and 41, by first introducing
the benzyloxy motif (42, 43) followed by transformation
of the carboxyl function into a nitrile (Scheme 3).18
Nitriles 25 and 26 were prepared similarly from (2-chlo-
ro-pyridin-4-yl)-methanol and (2-methyl-pyridin-4-yl)-
methanol, respectively. The carboxylic acid 42 can also
be degraded by a Curtius rearrangement into the corre-
sponding Boc-protected amine 44. This latter sequence
was found to be complementary to the one outlined
in Scheme 2 for the synthesis of compounds with N
substituents in the c-position of the pyridine ring. The
Boc-protected amine 44 proved to be a versatile building
block for the synthesis of many similar compounds, for
example, 18 (Table 4).
N
d
Cl
O
N
15
HO
Cl
e
f
O
CN CN
32
33
c
Br
N
N
Br
N
NH2
34
35
X
X
d
Cl
O
a
N
N
O
N
N
R
Cl
COOH
O
COOH
17
40: X = CH3
41: X = Cl
42: X = CH3
43: X = Cl
X
N
O
N+
N+
g
b, c
R
O2N
O
Cl
O
36
N
37
20: X = CH3
24: X = Cl
i-k
h
N
N
R
R
O
NH2
O
CN
d
e-g
38
39
N
42
R
O
NHBoc
O
c
N
44
Cl
O
N
N
16
Cl
O
N
CF3
Scheme 2. Reagents and conditions: (a) H2SO4, HNO3, 90 min 90 °C;
(b) Fe, CH3COOH, 1 h rt; (c) acetonylacetone, pTsOH Æ H2O, toluene,
9 h reflux (water trap) (9 h for 31 and 39, 1.5 h for 34); (d) 3-
chlorobenzyl alcohol, NaH, THF, 30 min rt, then 31/35, 12 h reflux; (e)
KCN, H2O, 10 °C, then 12 h rt; (f) HBr, acetic acid, 2 h rt; (g) 3-
chlorobenzyl alcohol, NaH, DMF, 30 min 60 °C, then 36 as a hot
solution in DMF, 2 h 60 °C; (h) dimethylcarbamoylchloride, TmsCN,
CHCl3, 1 h reflux; (i) NaOH, dioxane, H2O; (j) diphenylphosphoryl
azide, tBuOH, NEt3, 12 h reflux; (k) TFA, CH2Cl2, 4 days rt;
acetonylacetone, pTsOH Æ H2O, toluene, 24 h reflux (water trap).
18
Scheme 3. (R = 3-chlorobenzyl). Reagents and conditions: (a) 3-
chlorobenzyl alcohol, NaH, toluene, 30 min rt, then 40 (41), 18-
crown-6, 6h reflux (41: 12 h 55 °C); (b) 1,10-carbonyldiimidazole,
DMF, 30 min 60 °C (43: 60 min 45 °C), then NH4OH, 5 h rt; (c)
TFAA, NEt3, THF, 1 h rt; (d) diphenylphosphoryl azide, tBuOH,
NEt3, 12 h reflux; (e) NaH, CH3I, DMF, 1.5 h rt; (f) TFA, CH2Cl2,
12 h rt; (g) TFAA, NEt3, THF, 1 h rt.

1896
B. Bu¨ttelmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1892–1897
Table 6. Compound 20 and its analogues 24–26 with improved
molecular properties
4. Ballard, T. M.; Woolley, M. L.; Prinssen, E.; Huwyler, J.;
Porter, R.; Spooren, W. Psychopharmacology 2005, 179,
218.
5. Porter, R. H. P.; Jaeschke, G.; Spooren, W.; Ballard, T.;
Buettelmann, B.; Kolczewski, S.; Peters, J.-U.; Prinssen,
E.; Wichmann, J.; Vieira, E.; Muehlemann, A.; Gatti, S.;
Mutel, V.; Malherbe, P. J. Pharmacol. Exp. Ther. 2005,
315, 711.
6. (a) Woods, J. H.; Katz, J. L.; Winger, G. Pharmacol. Rev.
1992, 44, 151; (b) Vaswani, M.; Linda, F. K.; Ramesh, S.
Prog. Neuropsychopharmacol. Biol. Psychiatry 2003, 27,
85.
7. (a) Swanson, C. J.; Bures, M.; Johnson, M. P.; Linden,
A.-M.; Monn, J. A.; Schoepp, D. D. Nature Rev. Drug.
Discovery 2005, 42, 131; (b) Spooren, W.; Gasparini, F.
Drug News Perspectives 2004, 17, 251.
8. (a) Bradbury, M. J.; Campbell, U.; Giracello, D.; Chap-
man, D.; King, C.; Tehrani, L.; Cosford, N. D. P.;
Anderson, J.; Varney, M. A.; Strack, A. M. J. Pharmacol.
Exp. Ther. 2005, 313, 395; (b) Breysse, N.; Baunez, C.;
Spooren, W.; Gasparini, F.; Amalric, M. J. Neurosci.
2002, 22, 5669; (c) Zhu, C. Z.; Wilson, S. G.; Mikusa, J. P.;
Wismer, C. T.; Gauvin, D. M.; Lynch, J. L., III; Wade, C.
L.; Decker, M. W.; Honore, P. Eur. J. Pharmacol. 2004,
506, 107; (d) Mutel, V. Exp. Opin. Ther. Pat. 2002, 12,
1845; (e) Chiamulera, C.; Epping-Jordan, M. P.; Zocchi,
A.; Marcon, C.; Cottiny, C.; Tacconi, S.; Corsi, M.; Orzi,
F.; Conquet, F. Nat. Neurosci. 2001, 4, 873.
Ki = 1 nM
a
CLint = 218 ll/min/lg
Solubilityb < 1.0 lg/ml
logD7.4 > 3.0
N
20
24
Cl
O
CN
clogP = 3.6
Cl
Cl
Ki = 5 nM
CLinta = 47ll/min/lg
Solubilityb < 1.0 lg/ml
logD7.4 > 3.0
N
N
Cl
Cl
O
O
CN
CN
clogP = 3.8
Ki = 17 nM
CLinta = 20ll/min/lg
Solubilityb = 3.0 lg/ml
logD7.4 > 3.0
25
N
clogP = 2.4
Ki = 10 nM
Cl
CLinta = 37ll/min/lg
N
Solubilityb = 14 lg/ml
logD7.4 > 2.8
clogP = 2.1
26
O
CN
N
a From incubations with rat liver microsomes. lg = lg of protein.
b LYSA, for measurement details, see Ref. 11.
9. (a) Varney, M. A.; Cosford, N. D.; Jachec, C.; Rao, S. P.;
Sacaan, A.; Lin, F. F.; Bleicher, L.; Santori, E. M.; Flor,
P. J.; Allgeier, H.; Gasparini, F.; Kuhn, R. J. Pharmacol.
Exp. Ther. 1999, 290, 170; (b) Gasparini, F.; Lingenhohl,
K.; Stoehr, N.; Flor, P. J.; Heinrich, M.; Vranesic, I.;
Biollaz, M.; Allgeier, H.; Heckendorn, R.; Urwyler, S.;
Varney, M. A.; Johnson, E. C.; Hess, S. D.; Rao, S. P.;
Sacaan, A. I.; Santori, E. M.; Velicelebi, G.; Kuhn, R.
Neuropharmacology 1999, 38, 1493; (c) Cosford, N. D. P.;
Tehrani, L.; Roppe, J.; Schweiger, E.; Smith, N. D.;
Anderson, J.; Bristow, L.; Brodkin, J.; Jiang, X.; McDon-
ald, I.; Rao, S.; Washburn, M. J. Med. Chem. 2003, 46,
402.
In conclusion, 20 and 24–26 (Table 6) represent the best
compounds that were obtained during a medicinal
chemistry programme aiming for structurally novel
mGluR5 antagonists. Among these compounds, 20 has
the highest mGluR5 affinity. Compound 20 has been
characterised in an in vitro metabolite identification
study and in an in vivo PK study. Compounds 20 and
24–26 showed anxiolytic-like effects in an animal model
after oral administration.
10. Measurements were performed according to: Malherbe,
P.; Kratochwil, N.; Zenner, M.-T.; Piussi, J.; Diener, C.;
Kratzeisen, C.; Fischer, C.; Porter, R. H. P. Mol.
Pharmacol. 2003, 64, 823.
Acknowledgments
The excellent technical assistance of Antonio Ricci,
´
Beatrice David, Daniel Rueher, Ferhat Ucan, Regina
Wolf, Sandra Steiner, Silja Weber and Stephane Kritter
11. Gonzalez, R. C. B.; Huwyler, J.; Boess, F.; Walter, I.;
Bittner, B. Biopharm. Drug. Dispos. 2004, 25, 37.
12. Measurements were performed according to an in-house
developed method to assess solubility from a 10 mM
DMSO stock solution. This method is similar to the
classical thermodynamic shake-flask solubility with the
only difference that DMSO is removed in the beginning by
an additional lyophilization step. Therefore, this assay is
called LYophilisated Solubility Assay (LYSA).
13. Alagille, D.; Baldwin, R. M.; Roth, B. L.; Wroblewski, J.
T.; Grajkowska, E.; Tamagnan, G. D. Bioorg. Med. Chem.
Lett. 2005, 15, 945.
14. Alagille, D.; Baldwin, R. M.; Roth, B. L.; Wroblewski, J.
T.; Grajkowska, E.; Tamagnan, G. D. Bioorg. Med. Chem.
2005, 13, 197.
15. Vogel, J. R.; Beer, B.; Clody, D. E. Psychopharmacologia
1971, 21, 1.
´
is gratefully acknowledged.
References and notes
1. (a) Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F.
Pharmacol. Rev. 1999, 51, 7; (b) Pin, J.; De Colle, C.;
Bessis, A.; Acher, F. Eur. J. Pharmacol. 1999, 375, 277; (c)
O’Hara, L. J.; Sheppard, P. O.; Thogersen, H.; Venezia,
D.; Haldema, B. A.; MacGrane, V.; Houamed, K. M.;
Thomsen, C.; Gilbert, E. R.; Mulvihill, E. R. Neuron 1993,
11, 41.
2. (a) Conn, J. P.; Pin, J. P. Annu. Rev. Pharmacol. Toxicol.
1997, 37, 205; (b) Kno¨pfel, T.; Kuhn, R.; Allgeier, H. J.
Med. Chem. 1995, 38, 1417.
3. (a) Spooren, W. P. J. M.; Gasparini, F.; Salt, T. E.;
Kuhn, R. Trends Pharmacol. Sci. 2001, 22, 331; (b)
Shigemoto, R.; Nomura, S.; Ohishi, H.; Sugihara, H.;
Nakanishi, S.; Mizuno, N. Neurosci. Lett. 1993, 163,
53; (c) Romano, C.; Sesma, M. A.; McDonald, C. T.;
O’Malley, K.; VandenPol, A. N.; Olney, J. W. J.
Compar. Neurol. 1995, 355, 455.
16. Wessely, F.; Eibl, H.; Friedrich, G. Monatsh. Chem. 1952,
83, 24.
17. Adams, R.; Reifschneider, W. J. Am. Chem. Soc. 1957, 79,
2236.
18. Preparation of 2-(3-Chloro-benzyloxy)-6-methyl-isonico-
tinonitrile (20): (a) 2-(3-Chloro-benzyloxy)-6-methyl-isoni-
cotinic acid (42): NaH (60% in mineral oil, 5.42 g,

B. Bu¨ttelmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1892–1897
1897
226 mmol) was added in small portions to a mechanically
stirred suspension of 3-chlorobenzyl alcohol (9.97 g,
70 mmol) and 2-chloro-6-methylisonicotinic acid (40,
10 g, 58 mmol) in toluene (110 ml). After addition of 18-
crown-6 (1.3 g), the viscuous mixture was heated to 125 °C
(4 h). HCl (1 N, 70 ml), then heptane (250 ml) was added
to the cooled reaction mixture (ice bath), and the mixture
was stirred for 1 h (rt). The precipitate (12.14 g, 75%) was
After workup (EtOAc/H2O), drying (Na2SO4), and evap-
oration of the solvent, the title compound (2.2 g, 73%) was
obtained by recrystallisation from isopropyl ether/EtOAc.
1H NMR (DMSO-d6): d 2.44 (3H, s), 5.37 (2H, s), 7.07
(1H, s), 7.25 (1H, s), 7.38–7.43 (4H, m), 7.59 (1H, s), 7.59
(1H, br s), 8.07 (1H, br s); (c) 2-(3-Chloro-benzyloxy)-6-
methyl-isonicotinonitrile (20): Trifluoroacetic anhydride
(900 mg, 43 mmol) is added to a cooled (5 °C) mixture
of 2-(3-chloro-benzyloxy)-6-methyl-isonicotinamide (1 g,
3.61 mmol), CH2Cl2 (40 ml) and triethylamine (730 mg,
7.21 mmol). The reaction mixture was kept at rt (2 h)
before it was worked up (extraction EtOAc/brine), dried
(Na2SO4) and evaporated. The title compound (510 mg,
55%) was isolated from the residue by column chromatog-
raphy (silica gel, EtOAc/heptane = 2:3). 1H NMR (DMSO-
d6): d 2.46 (3H, s), 5.37 (2H, s), 7.25 (1H, s), 7.31 (1H, s),
7.39–7.43 (4H, m), 7.54 (1H, s). MS: m/e = 259.1 [M + H+].
Compounds 24–26 are prepared similarly.
1
filtered and dried at 50 °C. H NMR (DMSO-d6): d 2.46
(3H, s), 5.37 (2H, s), 7.07 (1H, s), 7.30 (1H, s), 7.39–7.44
(4H, m), 7.54 (1H, s), 13.56 (1H, br s); (b) 2-(3-Chloro-
benzyloxy)-6-methyl-isonicotinamide: 1,1-Carbonyldiimi-
dazole (2.1 g, 13 mmol) was added to a solution of 2-(3-
chloro-benzyloxy)-6-methyl-isonicotinic acid (42, 3 g,
10.8 mmol) in DMF (50 ml), and the mixture was heated
(60 °C) for 30 min. Upon cooling to rt, ammonia (25%,
30 ml) was added, and the mixture was stirred until the
reaction was complete, as indicated by TLC (ca. 2 h).