The acid-mediated intramolecular 1,3-dipolar cycloaddition of derived 2-nitro-1,1-ethenediamines for the synthesis of novel fused bicyclic isoxazoles

Article abstract of DOI:10.1016/j.tetlet.2010.04.098

The discovery of a novel synthesis of new fused bicyclic isoxazoles, for example, N-methyl-3-phenyl-5,6-dihydro-4H-isoxazolo[3,4-c]azepin-8-amine (2a), N-methyl-3-phenyl-4,5-dihydroisoxazolo[3,4-c]pyridin-7-amine (2b) and N-methyl-3-phenyl-4H-pyrrolo[3,4-

Full text of DOI:10.1016/j.tetlet.2010.04.098

Tetrahedron Letters 51 (2010) 3388–3391
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The acid-mediated intramolecular 1,3-dipolar cycloaddition of derived
2-nitro-1,1-ethenediamines for the synthesis of novel fused bicyclic isoxazoles
*
Lee W. Page , Matthew Bailey, Paul J. Beswick, Simon Frydrych, Robert J. Gleave
GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery of a novel synthesis of new fused bicyclic isoxazoles, for example, N-methyl-3-phenyl-5,6-
dihydro-4H-isoxazolo[3,4-c]azepin-8-amine (2a), N-methyl-3-phenyl-4,5-dihydroisoxazolo[3,4-c]pyri-
din-7-amine (2b) and N-methyl-3-phenyl-4H-pyrrolo[3,4-c]isoxazol-6-amine (2d) in high yield is
reported. We speculate that the reaction proceeds via acid-mediated intramolecular 1,3-dipolar cycload-
dition from 2-nitro-1,1-ethenediamines 1a,b,d.
Received 6 January 2010
Revised 1 April 2010
Accepted 23 April 2010
Available online 28 April 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
1,3-Dipolar cycloaddition
Nitrile oxide
2-Nitro-1,1-ethenediamine
Fused bicyclic isoxazole
7-Ring fused isoxazole
6-Ring fused isoxazole
The isoxazole ring is a commonly occurring structural fragment
in biologically active molecules, for example, natural products such
as muscimol and ibotenic acid1 and marketed drugs (e.g., Valdec-
oxib, BextraÒ).2 Thus isoxazoles have been widely used and studied
in modern drug discovery.3
cycloaddition onto the phenyl ring was observed. It is pertinent
to note that this is, to the best of our knowledge, the first example
N
Isoxazoles can be prepared in many ways, but the most widely
reported and researched synthesis of isoxazoles is via a [3+2]
cycloaddition of a nitrile oxide and an alkyne. Conventional gener-
ation of a nitrile oxide requires dehydration of a nitroalkane or
similar starting material.4–8 Other methods for generating nitrile
oxides are by reaction of aldoximes with oxidising agents9 or hal-
ogenating species10 (in the latter case the reaction proceeds via
hydroximoyl chlorides, followed by elimination of HCl with base).
Herein, we report the discovery and optimisation of a novel syn-
thesis of previously unreported fused bicyclic isoxazoles 2a,b,d
(Fig. 1). We believe that the reaction proceeds via acid-mediated
intramolecular 1,3-dipolar cycloaddition from 2-nitro-1,1-ethen-
ediamines 1a,b,d.
N
NH
NH
O
N
O N
N-methyl-3-phenyl-4,5-dihydroisoxazolo
[3,4-c]pyridin-7-amine 2b (83%)
N-methyl-3-phenyl-5,6-dihydro-4H-
isoxazolo[3,4-c]azepin-8-amine 2a (79%)
N
NH
O
N
N-methyl-3-phenyl-4H-pyrrolo
[3,4-c]isoxazol-6-amine 2d (~1%)
Figure 1. Novel bicyclic isoxazoles 2a,b,d with isolated yields.
Treatment of 2-nitro-1,1-ethenediamine derivative 1a with TFA
at 70 °C affords the bicyclic isoxazole heterocycle 2a in good yield
(Scheme 1). This procedure exploits the ability of the 2-nitro-1,1-
ethenediamine functionality to rearrange under acidic conditions.
Recently, Coustard et al.11 reported that similar intermediates
could generate nitrile oxides in situ which subsequently cyclise
intramolecularly onto a phenyl ring.12 With our intermediates no
O
N+
O
N
a
N
N
H
H
O
NH
N
1a
2a
* Corresponding author. Tel.: +44 1279 643179; fax: +44 1279 622005.
E-mail addresses: Lee.W.Page@GSK.com, Lee.W.Page@live.com (L.W. Page).
Scheme 1. Reagents and conditions: (a) TFA (10 equiv), MeCN, 70 °C, 24 h, followed
by rt 48 h.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.04.098

L. W. Page et al. / Tetrahedron Letters 51 (2010) 3388–3391
3389
Table 1
of the formation of an isoxazole via acid-mediated intramolecular
1,3-dipolar cycloaddition from 2-nitro-1,1-ethenediamines.
The procedure is versatile as the 2-nitro-1,1-ethenediamine
derivatives can be readily prepared from a wide variety of amines
using commercial reagents under mild conditions. Several syn-
thetic approaches to intermediates 1a–d were utilised. The route
shown in Scheme 2 afforded alcohols 4a–c via Sonogashira cou-
pling with commercial acetylene alcohol derivatives 3a–c.13
Amines 5a–c were prepared utilising Staudinger chemistry,14 and
subsequent treatment with N-methyl-1-(methylthio)-2-nitroe-
then-1-amine gave 2-nitro-1,1-ethenediamines 1a–c in high yield.
2-Nitro-1,1-ethenediamine 1d was prepared according to
Scheme 3. Commercial 2-propyn-1-amine (6) was first treated
with N-methyl-1-(methylthio)-2-nitroethen-1-amine to give the
acetylene-nitro-etheneamine derivative 7, and subsequent Sono-
gashira coupling afforded 1d.
To determine optimal conditions rapidly several trial reactions
against each variable were piloted. Each reaction was monitored
crudely by UV absorbance percentage peak area, which was con-
firmed by observation of the product molecular ion using LC-MS.15
The studies show that the use of the microwave irradiation
greatly reduces the reaction time whilst maintaining comparable
yields to those of conventional thermal conditions. For example,
cyclisation typically occurring in 2–3 h is comparable to 5 min
with microwave irradiation (Table 1).
An investigation on multiple variables to optimise the synthesis of bicyclic isoxazoles
2a,b,d
Substrate
Heat
Time (min)
T (°C)
Solvent
Producta (UV %)
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1d
1d
1d
D
D
D
D
30
60
120
180
5
5
5
5
5
83
83
83
MeCN
MeCN
MeCN
MeCN
MeCN
DMF
DMSO
THF
H2O
2a (31)
2a (52)
2a (79)
2a (86)
2a (78)
2a (57)
2a (57)
2a (21)
2a (10)
No product
2a (22)
2a (69)
2a (72)
2a (76)
2a (77)
2a (65)
2a (63)
2a (77)
2b (90)
2b (86)
2b (75)
2b (32)
2d (51)
2d (27)
2d (51)
83
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
VW
MW
MW
MW
MW
MW
VW
140
140
140
140
140
140
140
140
140
140
140
100
120
160
140
120
120
100
140
120
120
5
5
EtOH
CH2Cl2
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
0.5
2
10
20
5
5
5
5
10
15
10
5
5
10
The data in Table 1 show that polar aprotic solvents are optimal
solvents. Polar aprotic solvents are very efficient at absorbing
microwave irradiation and are also good at solvating cations. This
can lead to increased reactivity and improved reaction rates. Ace-
tonitrile was the highest yielding solvent. Less polar solvents such
as CH2Cl2 and THF performed poorly and some decomposition of
the starting material was observed. Polar protic solvents (EtOH,
water) were either very poor yielding or gave no product at all. It
is likely that the hydrogen bonding abilities of the polar protic
solvents could lead to solvent cage-like effects which stabilise
intermediates and prevent product formation; reversibility on
work-up could also explain the presence of starting materials in
these cases.
Each reaction was conducted with 2 equiv of concd H2SO4 present.
a
Results quoted as percentage area from the LC-MS UV trace.
Precise mechanistic details remain unclear. One suggested
mechanism by Coustard et al.11 uses a strong acid to generate a ni-
trile oxide in situ which would subsequently undergo 1,3-dipolar
cycloaddition onto the alkyne. It is equally plausible that the 1,3-
dipolar cycloaddition occurs via a stepwise process. Scheme 4
shows a stepwise mechanism proposed from similar work by Ohw-
ada et al.16 In this case, O-protonated aci-nitro species generated
in situ would undergo 1,3-dipolar cycloaddition. Further work is
required to determine the real pathway to the isoxazole.
Time was less critical on the outcome of the 1,3-dipolar cyclo-
addition reaction. The optimal time would appear to be 5 minutes
by the recorded UV % peak areas shown by LC-MS. However, as
long as the optimal temperature was reached, a high yield of prod-
uct was observed across the time ranges from 30 s to 20 min.
The optimal temperature was found to be 140 °C for forming
the 7-ring fused isoxazole 2a. It was also the case that 140 °C for
5 min was optimal for the formation of 6-ring fused isoxazole 2b
and the 5-ring fused isoxazole 2d (Table 1).
n
n OH
4a (n=3)
4b (n=2)
4c (n=4)
OH
a
3a (n=3)
3b (n=2)
3c (n=4)
O
N+
O
n
NH2
n
N
NH
b, c, d
e
H
5a (n=3)
5b (n=2)
5c (n=4)
1a (n=3)
1b (n=2)
1c (n=4)
X
H
H
N
Acid
N
Acid
n
n
1a,b,d
H
N
Scheme 2. Reagents and conditions: (a) iodobenzene, (Ph3P)2PdCl2, CuI, Et3N,
MeCN, 40 °C, 80–87%; (b) MsCl, Et3N, CH2Cl2, rt, 80–91%; (c) NaN3, DMF, 60 °C, 93–
99%; (d) Ph3P, THF, H2O, 60 °C, 70–78%; (e) N-methyl-1-(methylthio)-2-nitroethen-
1-amine, EtOH, H2O, rt, 76–82%.
NH+
O
+N
O
X
H+
O
H
+ N
O
H
X
H
H+
O
N+
O
N+
1,3-dipolar
addition of
aci-nitro
O
NH
X
+
nNH
O
nN
Heat/Energy
species
N
a
b
NH
X
H
NH
NH2
N
NH
H
-H2O
-2HX
H
O
N
6
O N
1d
7
OH2
2a (n=3)
2b (n=2)
2d (n=1)
+
Scheme 3. Reagents and conditions: (a) N-methyl-1-(methylthio)-2-nitroethen-1-
amine, EtOH, H2O, rt, 80%; (b) iodobenzene, (Ph3P)2PdCl2, CuI, Et3N, MeCN, 40 °C,
82%.
Scheme 4. Plausible mechanism for the formation of compounds 2a,b,d.

3390
L. W. Page et al. / Tetrahedron Letters 51 (2010) 3388–3391
Further studies (Table 2) indicated that two equivalents of con-
has been developed and the products characterised.18 The optimal
conditions for the intramolecular 1,3-dipolar cycloaddition reac-
tion on intermediates 1a,b,d are two equivalents of concentrated
sulfuric acid in acetonitrile (0.2 mol/L) with microwave irradiation
at 140 °C for 5 min.
This finding is a valuable addition to the literature of isoxazoles
formed by the [3+2] cycloaddition reaction. Our derived 2-nitro-
1,1-ethenediamines containing an intramolecular alkyne have
formed a new class of bicyclic isoxazole systems of which very lit-
tle is known. Moreover, this report forms a starting point for the
synthesis of further novel heterocycles using similar methodology.
centrated sulfuric acid (pKa À3)17 were the best combination of
acid and equivalents for the successful cycloadditions. Concen-
trated hydrochloric acid (pKa À8)17 and TFA (pKa 0.25)17 gave con-
siderable product peaks by LC-MS (UV %) but weaker acids such as
acetic acid (pKa 4.76)17 and phosphoric acid (pKa 2.12)17 resulted
in little or no products with mainly starting material recovered.
Interestingly, methanesulfonic acid (pKa À0.6)17 and trifluoro-
methanesulfonic acid (pKa À14)17 gave little product formation de-
spite being strong acids.
It is clear that the stoichiometric involvement of 2 equiv of acid
is optimal (Table 2). Too little acid prevents full consumption of all
the starting materials, but with too much acid, increased by-prod-
uct formation and decomposition were evident. The proposed
mechanism in Scheme 4 is supportive of the use of two stoichiom-
etric amounts of acid in order to achieve the isoxazole with an
overall loss of water from the starting materials.
Acknowledgement
Thanks to the BA crest award, Nuffield Science Bursaries, for the
financial support to fund this research.
It was not possible to prepare the novel bicyclic 8-ring fused
isoxazole from intermediate 2-nitro-1,1-ethenediamine 1c. We
propose that the greater ‘degrees of freedom’ within this system
(n = 4) make it less likely to achieve the required conformation
for cycloaddition, and in these cases, competing by-products are
formed in preference, even with longer times and elevated temper-
atures only decomposition products were observed. It was also not
possible to achieve intermolecular cycloaddition reactions with
external sources of alkyne. We assume that the protonated 2-ni-
tro-1,1-ethenediamine generated was prone to either intramolecu-
lar chemistry or decomposition before any external alkyne source
can get close enough to react (Table 2).
When using the optimised conditions of acetonitrile for 5 min of
microwave irradiation at 140 °C in the presence of 2 equiv of con-
centrated sulfuric acid, novel compounds 2a and 2b were isolated
in high yield (79% and 83%, respectively).18 The 5-ring fused isox-
azole 2d was isolated with good crude weight but after chromatog-
raphy, only poor yields of clean material were recorded.18
In summary a novel synthesis of previously unreported fused
bicyclic isoxazoles 2a (79%), 2b (83%) and 2d (poor yield/stability)
References and notes
1. Matsushima, Y.; Eguchi, F.; Kikukawa, T.; Matsuda, T. Inflamm. Regen. 2009, 29,
47–58.
2. Talley, J.; Brown, D.; Carter, J.; Graneto, M.; Koboldt, C.; Masferrer, J.; Perkins,
W.; Rogers, R.; Shaffer, A.; Zhang, Y.; Zweifel, B.; Seibert, K. J. Med. Chem. 2000,
43, 775–777.
3. (a) Wade, P. A.. In Comprehensive Organic Synthesis; Trost, B. M., Ed.; Pergamon
Press: Oxford, 1991; Vol. 3, p 1111; (b) Conti, P.; Dallanoce, C.; Amici, M. D.;
Micheli, C. D.; Klotz, K. N. Bioorg. Med. Chem. 1998, 6, 401–408; (c) Burkhart, D.
J.; Twamley, B.; Natale, N. R. Tetrahedron Lett. 2001, 42, 8415–8418; (d)
Srirastara, S.; Bajpai, L. K.; Batra, S.; Bhaduri, A. P.; Maikhurui, J. P.; Gupta, G.;
Dhar, J. D. Bioorg. Med. Chem. 1999, 7, 2607–2613.
4. Mukaiyama, T.; Hoshino, T. J. Am. Chem. Soc. 1960, 82, 5339–5342.
5. Kantorowski, E. J.; Brown, S. P.; Kurth, M. J. J. Org. Chem. 1998, 63, 5272–5274.
6. (a) Basel, Y.; Hassner, A. Synthesis 1997, 309–312; (b) Zagozda, M.; Plenkiewicz,
J. Tetrahedron: Asymmetry 2007, 18, 1457–1464.
7. Shimizu, T.; Hayashi, Y.; Shibafuchi, H.; Teramura, K. Bull. Chem. Soc. Jpn. 1986,
59, 2827.
8. Gao, S.; Tu, Z.; Kuo, C.-W.; Liu, J.-T.; Chu, C.-M.; Yao, C.-F. Org. Biomol. Chem.
2006, , 4, 2851–2857.
9. Just, G.; Dahl, L. Tetrahedron 1968, 24, 5251–5269.
10. (a) Grundmann, C.; Dean, J. M. J. Org. Chem. 1965, 30, 2809–2812; (b) Hassner,
H.; Rai, K. M. L. Synthesis 1989, 57–59; (c) Kim, J. N.; Ryn, E. K. Synth. Commun.
1990, 20, 1373.
11. (a) Soro, Y.; Bamba, F.; Siaka, S.; Coustard, J.-M. Tetrahedron Lett. 2006, 47,
3315–3319; (b) Coustard, J.-M. Tetrahedron 1995, 51, 10929–10940; (c)
Coustard, J.-M. Tetrahedron 1996, 52, 9505–9520; (d) Coustard, J.-M.
Tetrahedron 1999, 55, 5809–5820; (e) Cousson, A.; Coustard, J.-M.
Tetrahedron 1998, 54, 6523–6528; (f) Coustard, J.-M. Eur. J. Org. Chem. 2001,
1525–1531.
Table 2
The effects of acid and equivalents on the microwave-assisted synthesis of bicyclic
isoxazoles 2a,b,d
12. Soro, Y.; Bamba, F.; Siaka, S.; Coustard, J.-M.; Adima, A. A. J. Chem. Sci. 2007, 119,
259–265.
Substrate
Acid
Equivalents
Producta (UV %)
13. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 50, 4467.
14. (a) Naguma, S.; Kawahara, N.; Bando, H.; Imai, M.; Higuchi, T.; Saito, H.;
Mizukami, M. Tetrahedron Lett. 2007, 48, 7228; (b) Campi, E. M.; Chong, M. J.;
Jackson, R. W.; Van der Schoot, M. Tetrahedron 1994, 50, 2533–2542; (c) Taylor,
E. C.; Pont, J. L. Tetrahedron Lett. 1987, 28, 379.
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
1c
1cc
1cd
1ce
1d
H2SO4
H2SO4
H2SO4
H2SO4
H2SO4
TFA
2
1
0.5
3
5
2
2
2
4
2
2
2
4
2
2
1
2
10
2
2a (78)
2a (67)
2a (59)
2a (72)
2a (64)
2a (69)
2a (69)
2a (1)
15. Electrospray ionisation technique with quadrupole mass detector (WatersÒ
,
Acquity™). Formic acid generic analytic UPLC open access liquid
chromatography mass spectra (LC-MS) 2 min method; LC conditions; The
UPLC analysis was conducted on an Acquity UPLC BEH C18 column
HCl
MeSO3H
MeSO3H
CF3SO3H
H3PO4
AcOH
No product
2a (11)
(50 Â 2.1 mm id 1.7 lm packing diameter) at 40 °C, the solvents employed
were A = 0.1% v/v solution of formic acid in H2O and B = 0.1% v/v solution of
formic acid in MeCN, the gradient employed was 0 min 97% A–3% B; 1.5 min 0%
A–100% B; 1.9 min 0% A–100% B; 2.0 min 97% A–3% B, flow rate 1 ml/min, the
UV detection was a summed signal from wavelength of 210 to 350 nm, MS
conditions; MS Waters ZQ, Ionisation mode; alternative-scan positive and
negative electrospray, Scan range; 100–1000 AMU, Scan time; 0.27 s, Inter scan
delay 0.10 s.
No productb
2a (10)
AcOH
2a (6)
2b (90)
H2SO4
H2SO4
H2SO4
H2SO4
H2SO4
H2SO4
H2SO4
No product
No product
No product
No product
No product
2d (51)
16. (a) Ohwada, T.; Okabe, K.; Ohta, T.; Shudo, K. Tetrahedron 1990, 46, 7539–7555;
(b) Nakamura, S.; Sugimoto, H.; Ohwada, T. J. Am. Chem. Soc. 2007, 129, 1724–
1732.
17. (a) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456; (b) Brownstein, S.; Stillman, A.
E. J. Phys. Chem. 1959, 63, 2061; (c) Dippy, J. F. J.; Hughes, S. R. C.; Rozanski, A. J.
Chem. Soc. 1959, 2492.
2
Each reaction was conducted at 140 °C for 5 min in the microwave oven with MeCN
as solvent.
18. All reactions were performed with
microwave reaction vessel. General procedure for the synthesis of bicyclic
isoxazoles: To suspension of nitro-etheneamine derivative (1a–d)
a Biotage initiator (2.5) in a sealed
a
Results quoted as percentage area from the LC-MS UV trace.
Poor solubility of the acid in the reaction mixture.
Intermolecular reaction attempted with benzylethyne (5 equiv).
Intermolecular reaction attempted with benzylethyne (4 equiv).
a
b
(0.815 mmol) in MeCN (4 mL) was added concd H2SO4 (2 equiv) (1.63 mmol).
The mixture was then submitted to microwave irradiation for 5 min at 140 °C.
After cooling, the reaction solvent was evaporated in vacuo and the crude
sample was purified by flash chromatography with a gradient of 100% CH2Cl2
c
d
e
Intermolecular reaction attempted with propynoic acid ethyl ester (5 equiv).

L. W. Page et al. / Tetrahedron Letters 51 (2010) 3388–3391
3391
to 10% (2 M NH3 in MeOH) in CH2Cl2. Fractions containing clean product were
evaporated in vacuo and the product was dried in a vacuum oven at 50 °C
overnight.
found [M+H]+ 228.1134; 1H NMR (400 MHz, CDCl3): d 2.91 (t, J = 6.8 Hz, 2H),
2.99 (s, 3H), 3.78 (t, J = 6.7 Hz, 2H), 4.96–5.45 (br s, 1H NH), 7.38–7.63 (m, 3H),
7.75–7.79 (m, 2H); 13C NMR (100 MHz, CDCl3): d 19.8, 27.3, 46.1, 111.1, 126.0,
N-Methyl-3-phenyl-5,6-dihydro-4H-isoxazolo[3,4-c]azepin-8-amine
(2a):
127.8, 129.1, 130.0, 149.8, 152.1, 163.2; IR (solid):
m 3204, 2944, 1642, 1613,
White solid, mp 113–115 °C (dec); LC-MS [M+H]+ 242; HR-MS (ES+), calcd
[M+H]+ C14H16N3O, 242.1293, found [M+H]+ 242.1286; 1H NMR (400 MHz,
CDCl3): d 2.01–2.10 (m, 2H), 2.82–3.05 (m, 5H), 3.68–3.70 (m, 2H), 7.36–7.60
(m, 3H), 7.72–7.76 (m, 2H); 13C NMR (100 MHz, CDCl3): d 23.5, 27.7, 28.6, 47.9,
1555, 1449, 1166 cmÀ1
.
N-Methyl-3-phenyl-4H-pyrrolo[3,4-c]isoxazol-6-amine (2d): Pale brown solid,
mp 76–78 °C (dec); LC-MS [M+H]+ 214; HR-MS (ES+), calcd [M+H]+ C12H12N3O,
214.0980, found [M+H]+ 214.0976; 1H NMR (400 MHz, CDCl3): d 3.16 (s, 3H),
4.68 (s, 2H), 7.44–7.53 (m, 3H), 7.70–7.74 (m, 2H); 13C NMR (100 MHz, CDCl3):
113.6, 126.7, 127.7, 129.0, 130.1, 151.8, 157.0, 166.1; IR (solid):
m 3393, 3056,
2896, 1634, 1526, 1448, 1208, 1162 cmÀ1 N-Methyl-3-phenyl-4,5-
dihydroisoxazolo[3,4-c]pyridin-7-amine (2b): White solid, mp 145–147 °C
(dec); LC-MS [M+H]+ 228; HR-MS (ES+), calcd [M+H]+ C13H14N3O, 228.1137,
d 29.6, 49.7, 123.7, 126.3, 127.0, 129.2, 130.3, 155.0, 160.6, 167.8; IR (solid):
m
3253, 2931, 1744, 1621, 1488, 1450, 1411, 1166 cmÀ1
.