Synthesis of spirocyclic pyridoazepines as analogues of galanthamine by nucleophilic aromatic substitution of 3-substituted 2-chloropyridines

Article abstract of DOI:10.1002/ejoc.200700614

In this report we describe the synthesis of spirocyclic pyridoazepines starting from easily available precursors. The key step of our synthesis is an intramolecular nucleophilic aromatic substitution of the appropriate 3-substituted 2-chloropyridines. The final compounds, designed as simplified analogues of the alkaloid galanthamine, showed significant acetylcholinesterase inhibition activity. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200700614

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
SHORT COMMUNICATION
DOI: 10.1002/ejoc.200700614
Synthesis of Spirocyclic Pyridoazepines as Analogues of Galanthamine by
Nucleophilic Aromatic Substitution of 3-Substituted 2-Chloropyridines
Sofie Vanlaer,[a] Wim M. De Borggraeve,[a] and Frans Compernolle*[a]
Keywords: Spiro compounds / Nucleophilic aromatic substitution / Nitrogen heterocycles / Acetylcholinesterase / Amines
In this report we describe the synthesis of spirocyclic pyr-
idoazepines starting from easily available precursors. The
key step of our synthesis is an intramolecular nucleophilic
aromatic substitution of the appropriate 3-substituted 2-chlo-
alogues of the alkaloid galanthamine, showed significant
acetylcholinesterase inhibition activity.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
ropyridines. The final compounds, designed as simplified an- Germany, 2007)
Introduction
Galanthamine (GAL; 1), an alkaloid isolated from the
Amaryllidaceae species, is a centrally acting, selective, com-
petitive and reversible inhibitor of the neurotransmitter ace-
tylcholinesterase (AChE).[1] Symptomatic treatment for
Alzheimer’s disease (AD) involves restoring the acetylcho-
line levels in patients by means of inhibitors of AChE. In
this respect, galanthamine significantly enhances cognitive
functions in AD patients,[2] and recently it was approved in
Europe and in the USA for treatment of AD. Galanthamine
can be isolated from botanical sources but these sources are
limited. A few total syntheses of GAL,[3] and the synthesis
of various analogues,[4] have been reported. Galanthamine
is structurally related to morphine (Figure 1). Morphine
and some simplified structural analogues are used for the
treatment of severe pain. Therefore, compounds containing
a simplified galanthamine skeleton might be of interest for
the development of both new inhibitors of acetylcho-
linesterase and further analogues of morphine.
Our goal is to make simplified analogues of GAL in
which the benzene ring is replaced with a pyridine unit.
Figure 1. (a) GAL, targets 2a–d, GAL analogue 3 and morphine;
Thus, whereas our pyridine target structures of type 2 retain
the spiroannulation, the ether linkage between the aromatic
ring and the spiro ring is disconnected (Figure 1a). In this
work, we present a synthetic route towards spirocyclic pyri-
doazepines 2a–d starting from easily available precursors.
Pyridoazepines are rather unknown, and to the best of our
knowledge, pyridoazepine ring systems encompassing a spi-
roannulation have not yet been reported. Compound 3, a
synthetic analogue of GAL that also lacks this heterocyclic
connection, was claimed to have the desired, albeit weak,
activity.[5]
(b) Geometrically optimised structures of Gal, 2c and 2d, and over-
lay of GAL and targets 2c and 2d by using Hyperchem MM+ mo-
lecular mechanics.
An overlay of the geometrically optimised model struc-
tures of GAL and targets 2c and 2d reveals an excellent fit
of the superimposed benzo- and pyridoazepine ring moie-
ties and of the corresponding spiro ring moieties (Fig-
ure 1b). The chlorine in the pyridine nucleus of 2a–d was
meant to mimic the methoxy group of GAL and in future
work may be substituted by various nucleophiles. The
amide oxygen atoms in the enantiomeric forms of 2a–c
(Figure 1) could serve as a mimic for the ether bridge of
GAL. Alternatively, the planar amide bond present in the
[a] Chemistry Department, K. U. Leuven,
Celestijnenlaan 200F, 3001 Heverlee, Belgium
Fax: +32-16-327990
E-mail: Frans.Compernolle@chem.kuleuven.be
Eur. J. Org. Chem. 2007, 4995–4998
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4995

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
S. Vanlaer, W. M. De Borggraeve, F. Compernolle
SHORT COMMUNICATION
image forms of 2a–c can be superimposed on the alkenic ozonolysis (the ozonide intermediate was cleaved by treat-
linkage of the cyclohexene ring (not shown in Figure 1b).
ment with NEt3).[7] Final reductive amination of aldehydes
9a,b with amine 6 by using NaCNBH3 as a reducing agent
afforded precursors 10a,b in good yields.[8]
Results and Discussion
The crucial and most difficult step in this reaction se-
quence was the final ring closure of precursors 10a,b by
nucleophilic aromatic substitution (NAS) to form the spiro-
annulated seven-membered ring products 2a,b. Indeed, gen-
eration of the quaternary carbon centre of the spirocycle
requires attack of a sterically encumbered anion at the
2-chloro position. This conversion could, however,
be achieved by applying microwave irradiation with
KN(SiMe3)2 as a base in toluene.[9,10] Desired products 2a,b
were not detected with the use of conventional heating con-
ditions. Presumably, this was due to decomposition of the
desired product when elevated temperatures or prolonged
reaction times were applied. The temperature and reaction
time conditions were also of critical importance in the mi-
crowave reaction. The best conditions to effect spirocyclis-
ation of 10a consisted of irradiation of the reaction mixture
at 100 °C for 15 min, which provided target compound 2a
in 85% yield after flash column chromatography.[11] In con-
trast to the straightforward synthesis of 2a, the transforma-
tion of six-membered lactam analogue 10b into correspond-
ing spirocyclic product 2b proved to be erratic. Ring closure
was best carried out by irradiating the reaction mixture at
80 °C for 10 min.[12] However, numerous side products were
formed and following purification by reverse-phase HPLC,
2b was isolated in poor yield (ca. 5%).[11]
Our first synthetic approach to form the desired spirocy-
clic pyridoazepines was based on intramolecular ring clo-
sure of the appropriate 3-substituted 2-chloropyridines car-
rying a nucleophilic lactam enolate entity (Scheme 1). The
latter pyridine precursors in their turn can be constructed
by reductive coupling of the corresponding secondary
amine and lactam aldehyde derivatives.
Scheme 1. Retrosynthetic approach to target compounds 2a,b.
The synthesis of the substituted pyridine component
started with the cyclocondensation of lactonitrile and oxalyl
chloride to form 3,5-dichloro-2H-1,4-oxazin-2-one (4),
which was subjected to a Diels–Alder reaction with propar-
gyl bromide (Scheme 2).[6] This cycloaddition (followed by
the elimination of CO2) is 100% regioselective and only 3-
(bromomethyl)pyridine 5 is formed. Subsequent amination
afforded corresponding 3-(methylaminomethyl)pyridine 6.
In view of our failure to produce six-membered lactam
target compound 2b in satisfactory yield, we envisaged the
alternative approach outlined in Scheme 3. Instead of form-
ing the spirocyclic ring system by NAS in the final ring
closure step, spirocyclic pyridoazepines 2c,d were con-
structed by manipulation of an appropriate bicyclic pyr-
idoazepine. This key intermediate in turn can be formed by
intramolecular NAS of the corresponding 3-[(4-methoxy-
4-oxobutyl)(methyl)amino]-substituted
2-chloropyridine.
Consequently, the critical NAS ring-closure step involved a
less-hindered precursor than that used in our first ap-
proach.
Scheme 2. Reagents and conditions: (a) lactonitrile, oxalyl chloride
(3 equiv.), Et3N·HCl (0.1 equiv.), chlorobenzene, 90 °C, 3 h (90%);
(b) propargyl bromide (2 equiv.), toluene, 70 °C, overnight (92%);
(c) NH2CH3 (5 equiv.), MeOH, room temp., 4 h (85%); (d) (i) LDA
(1 equiv.), THF, –78 °C, 10 min. (ii) allyl bromide (1 equiv.), –78 °C,
30 min. (97%); (e) cat. OsO4, NaIO4 (2 equiv.), Et2O, H2O, room
temp., 12 h (99%); (f) (i) O3, CH2Cl2, MeOH, –78 °C, 30 min (ii)
NEt3, –78 °C to room temp., overnight (the crude compound was
used immediately in the next step); (g) amine 9 (1 equiv.), MeOH,
acetic acid (pH = 6), NaCNBH3 (1 equiv.), aldehyde 9a,b (1 equiv.),
room temp., 15 min (85% over 2 steps); (h) KN(SiMe3)2 (1 equiv.),
toluene, microwave irradiation (2a, 85%).
Scheme 3. Retrosynthetic approach to 2c,d.
Substitution of 3-(bromomethyl)pyridine 5 with methyl
4-(methylamino)butanoate afforded amine 11, which
smoothly underwent internal NAS to form pyridoazepine
The aldehyde component was prepared by initial α al- 12 under conventional heating conditions (Scheme 4).[13]
lylation of commercial lactams 7a,b to give compounds Two equivalents of base were required to effect complete
8a,b. Subsequent oxidative cleavage of the allyl group conversion into compound 12, which was generated as the
yielded aldehydes 9a,b; this conversion was effected either corresponding anion. The latter turned out to be very sensi-
by overnight reaction with NaIO4 and catalytic OsO4 or by tive to oxidation; hence, air must be rigorously excluded.
4996
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4995–4998

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Spirocyclic Pyridoazepines as Analogues of Galanthamine
Final workup of the reaction mixture was carried out by each case, the critical reaction step was ring closure of the
cooling down to –78 °C, followed by the careful addition of 3-substituted 2-chloropyridine precursors, which proceeded
a saturated aqueous solution of NH4Cl.[14]
through intramolecular nucleophilic aromatic substitution.
In the first route, this NAS reaction directly afforded the
desired spirocyclic lactam compounds. However, this
method gave unsatisfactory results for the synthesis of six-
membered spirocyclic lactams. In the second route proceed-
ing via an early NAS step, bicyclic pyridoazepine ester 12
was generated as the key intermediate and the spirolactam
ring was constructed later on. Our target compounds 2a–d
showed significant AChE inhibition activity.
Acknowledgments
We thank Professor S. Toppet and K. Duerinckx for their assist-
ance with the NMR spectroscopic analysis, Ir. B. Demarsin for
HRMS measurements and D. Henot for preparative HPLC. We
thank Prof. C. Gielens for assistance with the AChE inhibition
tests. S. V. thanks the Institute for the Promotion of Innovation
through Science and Technology in Flanders (Belgium, I. W. T.)
and W. M. D. B. (Postdoctoral Fellow of the FWO-Flanders)
thanks the Fund for Scientific Research-Flanders (Belgium,
F. W. O.) for the fellowships received.
Scheme 4. Reagents and conditions: (a) NEt3 (3 equiv.), MeOH,
room temp., 4 h (85%); (b) (i) KN(SiMe3)2 (2.2 equiv.), toluene,
80 °C, 10 min. (ii) –78 °C, saturated NH4Cl, 5 min. (90%); (c) (i)
KOtBu (1.2 equiv.), THF, room temp., 10 min (ii) acrylonitrile
(1.2 equiv.), tBuOH, room temp., 15 min. (61%); (d) CoCl2
(2 equiv.), NaBH4 (10 equiv.), MeOH, room temp., overnight; (e)
MeOH, overnight, reflux (57% over 2steps); (f) CH3SO3H, toluene,
100 °C, 2 h. (83%); (g) (i) KOtBu (1.2 equiv.), tBuOH, room temp.,
1 h (ii) NH4Cl, H2O (60%).
[1] a) H. A. M. Mucke, Drugs Today 1997, 33, 251–257; b) M.
Weinstock, CNS Drugs 1999, 12, 307–323; c) M. Rainer, Drugs
Today 1997, 33, 273–279; d) H. M. Greenblath, G. Kryger, T.
Lewis, I. Silman, J. L. Sussman, FEBS Lett. 1999, 463, 321–
326.
[2] M. Colombres, J. P. Sagal, N. C. Inestrosa, Curr. Pharm. Des.
2004, 10, 3121–3130.
[3] a) B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 2000, 122,
11262–11263; b) B. M. Trost, W. Tang, F. D. Toste, J. Am.
Chem. Soc. 2005, 127, 14785–14803; c) J. Marco-Contelles, M.
Carreiras, C. Rodriguez, M. Villarroya, A. G. Garcia, Chem.
Rev. 2006, 106, 116–133; d) S. E. Gibson, R. J. Middleton,
Contemp. Org. Synth. 1996, 3, 447–471.
[4] a) A. H. Lewin, J. Szewczyk, J. W. Wilson, F. I. Carroll, Tetra-
hedron 2005, 61, 7144–7152; b) S. Y. Han, J. E. Sweeney, E. S.
Bachean, E. J. Schweiger, G. Porloni, J. T. Coyle, B. M. Davis,
M. M. Jouillé, Eur. J. Med. Chem. 1992, 27, 673–687.
[5] a) P. Liang, J. Liu, L. Hsin, C. Cheng, Tetrahedron 2004, 60,
11655–11660; b) P. Liang, L. Hsin, S. Pong, C. Hsu, C. Cheng,
J. Chin. Chem. Soc. 2003, 50, 449–456.
To construct the spirolactam ring of target compounds
2c,d, we envisaged initial Michael reaction of pyridoazepine
12 with either acrylonitrile, methyl acrylate or acrylamide.
However, because of the unreactive character and steric
crowding of the stabilised ester enolate anion of 12, the ad-
dition reaction only succeeded with acrylonitrile.[15] Selec-
tive reduction of resulting nitrile adduct 13 with NaBH4 in
the presence of CoCl2 furnished primary amine 14, which
underwent ring closure in refluxing methanol to provide
target product 2c.[16] Nitrile adduct 13 also could be con-
verted into corresponding amide 15 by heating with meth-
anesulfonic acid in toluene. Final ring closure was effected
by reaction with KOtBu in tBuOH at room temperature to
form the cyclic imide salt; acidic workup furnished target
product 2d.[17]
Compounds 2a–d were tested for their inhibition activity
on acetylcholinesterase relative to that of galanthamine.
Tests were performed by using Ellman’s colorimetric
method.[18] The KM measured was 186 µ. The tested com-
pounds showed significant AChE inhibition activity (KI =
514 µ for 2a; KI = 70 µ for 2b; KI = 99 µ for 2c; KI =
150 µ for 2d; KM = 186 µ), but lower than that of galan-
thamine (KI = 3 µ). From these data it clearly appears
that six-membered lactam compounds 2b,c exhibit superior
activity relative to that of five-membered lactam 2a.
[6] a) P. R. Carly, F. Compernolle, G. J. Hoornaert, Tetrahedron
Lett. 1995, 36, 2113–2116; b) L. Meerpoel, G. Deroover, K. J.
Van Aken, G. Lux, G. J. Hoornaert, Synthesis 1991, 765–768;
c) L. Meerpoel, G. J. Hoornaert, Synthesis 1990, 905–908.
[7] a) J. S. Hon, S. W. Lin, Y. W. Chen, Synth. Commun. 1993, 23,
1543; b) M. J. Aurell, L. Ceita, R. Mestres, A. Tortajada, Tetra-
hedron 1997, 53, 10883–10898.
[8] Spectroscopic data for compound 10a: 1H NMR (400 MHz,
CDCl3): δ = 7.69 (s, 1 H), 3.54 (s, 2 H), 3.29 (dd, J = 8.4,
5.4 Hz, 2 H), 2.85 (s, 3 H), 2.55 (t, J = 7.5 Hz, 2 H), 2.51–2.45
(m, 1 H), 2.36 (s, 3 H), 2.24 (s, 3 H), 2.18–2.12 (m, 2 H), 1.72–
1.58 (m,1 H), 1.57–1.45 (m, 1 H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 176.9, 148.6, 147.0, 142.5, 132.8, 131.8, 57.9, 55.9,
48.0, 42.6, 40.0, 30.1, 29.6, 25.5, 19.3 ppm. HRMS: calcd. for
C15H21Cl2N3O 329.1062; found 329.1063. Spectroscopic data
1
for compound 10b: H NMR (400 MHz, CDCl3): δ = 7.62 (s,
1 H), 3.46 (s, 2 H), 3.28–3.18 (m, 2 H), 2.92 (s, 3 H), 2.52–2.42
(m, 2 H), 2.35–2.28 (m, 1 H), 2.26 (s, 3 H), 2.18 (s, 3 H), 1.90–
1.80 (m, 2 H), 1.76–1.66 (m, 1 H), 1.55–1.40 (m, 3 H) ppm.
13C NMR (100 MHz, CDCl3): δ = 172.9, 148.4, 146.9, 142.7,
132.8, 131.8, 57.7, 55.8, 50.4, 42.4, 39.6, 35.3, 29.7, 27.1, 22.2,
Conclusions
Two synthetic routes towards spirocyclic pyridoazepines
were developed starting from easily available precursors. In
Eur. J. Org. Chem. 2007, 4995–4998
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4997

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
S. Vanlaer, W. M. De Borggraeve, F. Compernolle
SHORT COMMUNICATION
19.2 ppm. HRMS: calcd. for C16H23Cl2N3O 343.1218; found
343.1210.
[9] CEM-Discover, CEM Corporation P. O. Box 200 Matthews,
NC 28106.
KN(SiMe3)2 (0.5  in toluene, 14.4 mL, 7.21 mmol) under an
atmosphere of argon. The reaction mixture was stirred at 80 °C
for 10 min, and then it was cooled to –78 °C and a saturated
aqueous solution of NH4Cl (10 mL) was added. After 10 min,
the mixture was warmed to room temp. A saturated aqueous
solution of K2CO3 was added until pH 8. After extraction with
CH2Cl2 the organic layer was dried with MgSO4.
[10] Optimised procedure for NAS of 10a: To a solution of com-
pound 10a (10 mg, 0.03 mmol) in dry toluene was added potas-
sium bis(trimethylsilyl)amide (0.5  in toluene, 0.06 mL,
0.03 mmol). The reaction mixture was flushed with argon, and
the reaction vessel was closed. The mixture was irradiated in
the microwave apparatus at 100 °C for 15 min (200 W, no si-
multaneous cooling). After the addition of water, the mixture
was extracted with CH2Cl2. The organic layer was dried with
MgSO4 and concentrated under reduced pressure.
[15] Spectroscopic data for compound 13: 1H NMR (400 MHz,
CDCl3): δ = 7.25 (s, 1 H), 3.70 (s, 3 H), 3.66 (d, J = 7.5 Hz, 1
H), 3.51 (d, J = 7.5 Hz, 1 H), 2.87–2.83 (m, 2 H), 2.62–2.59
(m, 3 H), 2.34–2.24 (m, 2 H), 2.30 (s, 3 H), 2.29 (s, 3 H), 1.90–
1.80 (m, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 173.5,
155.8 147.8, 141.3, 133.3, 130.6, 120.2, 59.8, 55.1, 54.6, 52.2,
44.6, 34.5, 33.9, 18.8, 13.5 ppm. HRMS: calcd. for
C16H20ClN3O2 321.1244; found 321.1249.
[11] Spectroscopic data for compound 2a: 1H NMR (400 MHz,
CDCl3): δ = 7.24 (s, 1 H), 4.20 (d, J = 14.6 Hz, 1 H), 3.62 (d,
J = 14.6 Hz, 1 H), 3.46 (dd, J = 16.3, 9 Hz, 1 H), 3.30 (td, J [16] Spectroscopic data for compound 2c: 1H NMR
= 9, 3.4 Hz, 1 H), 3.11–3.05 (m, 1 H), 3.02–2.95 (m, 1 H), 2.92
(s, 3 H), 2.73–2.65 (m,1 H), 2.47 (s, 3 H), 2.48–2.40 (m, 1 H),
2.29 (s, 3 H), 2.09 (dt, J = 13, 8 Hz, 1 H), 1.85–1.80 (m, 1 H)
ppm. 13C NMR (100 MHz CDCl3): δ = 176.6, 158.8, 148.5,
141.9, 130.1, 58.2, 55.1, 54.0, 46.8, 43.9, 32.3, 32.1, 30.1,
18.8 ppm. HRMS: calcd. for C15H20ClN3O 293.1295; found
293.1297. Spectroscopic data for compound 2b: HRMS: calcd.
for C16H22ClN3O 307.1451; found 307.1448.
(300 MHz,CD3OD): δ = 8.47 (br. s, 1 H), 7.64 (s, 1 H), 4.45
(d, J = 12.3 Hz, 1 H), 4.19 (d, J = 12.3 Hz, 1 H), 3.65–3.35 (m,
3 H), 3.18–3.10 (m, 1 H), 2.77 (s, 3 H), 2.68–2.50 (m, 2 H),
2.37 (s, 3 H), 2.25–2.19 (m, 1 H), 2.08–1.96 (m, 1 H), 1.94–1.84
(m, 2 H) ppm. 13C NMR (100 MHz, DMSO): δ = 173.6, 161.4,
146.6, 141.9, 132.7, 128.7, 56.3, 52.7, 52.6, 44.4, 41.5, 32.0,
31.8, 18.2, 18.1 ppm. HRMS: calcd. for C15H20ClN3O
293.1295; found 293.1293.
[12] This ring closure was carried out at 80 °C for 10 min instead
of 100 °C for 15 min. At 100 °C, compound 2b could not be
detected.
[17] Spectroscopic data for compound 2d: 1H NMR (400 MHz,
CD3OD): δ = 8.43 (br. s, 1 H), 7.65 (s, 1 H), 4.24 (d, J =
14.6 Hz, 1 H), 3.99 (d, J = 14.6 Hz, 1 H), 3.36–3.27 (m, 1 H),
3.18–3.12 (m, 1 H), 2.82–2.68 (m, 2 H), 2.61 (s, 3 H), 2.60–2.54
(m, 1 H), 2.39 (s, 3 H), 2.44–2.28 (m, 1 H), 2.24–2.16 (m, 1 H),
2.04–1.96 (m, 1 H) ppm. 13C NMR (100 MHz, CD3OD): δ =
178.7, 176.1, 159.7, 150.6, 145.2, 133.3, 124.7, 58.5, 55.0, 54.8,
44.7, 33.2, 30.1, 29.7, 19.4 ppm. HRMS: calcd. for
C15H18ClN3O2 307.1088; found 307.1097.
[13] Spectroscopic data for compound 12: 1H NMR (300 MHz,
CDCl3): δ = 7.32 (s, 1 H), 4.15 (dd, J = 6.8, 2.8 Hz, 1 H), 3.80
(d, J = 14.8 Hz, 1 H), 3.75 (s, 3 H), 3.59 (d, J = 14.8 Hz, 1 H),
3.08–2.90 (m, 2 H), 2.34 (s, 3 H), 2.33 (s, 3 H), 2.24–2.19 (m,
1 H), 2.12–2.06 (m, 1 H) ppm. 13C NMR (75 MHz, CDCl3): δ
= 172.4, 156.5, 148.4, 141.4, 133.5, 130.5, 59.7, 56.8, 53.0, 52.1,
43.7, 26.2, 18.9 ppm. HRMS: calcd. for C13H17ClN2O2
268.0978; found 268.0971.
[18] G. L. Ellman, K. D. Courtney, V. Andres, R. M. Featherstone,
Biochem. Pharmacol. 1961, 7, 88–95.
[14] Optimised procedure for NAS of 11: To a solution of pyridine
Received: July 4, 2007
Published Online: September 10, 2007
11 (1 gram, 3.28 mmol) in dry toluene (50 mL) was added
4998
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 4995–4998