Synthesis of isoxazoles en route to semi-aromatized polyketides: Dehydrogenation of benzonitrile oxide-para-quinone acetal cycloadducts

Article abstract of DOI:10.1039/c2ob25423a

A variety of highly functionalized polycyclic isoxazoles are prepared by a two-step protocol: (1) 1,3-dipolar cycloaddition of o,o′-disubstituted benzonitrile oxides to para-quinone mono-acetals, then (2) dehydrogenation. The cycloaddition proceeds in a regioselective manner, favouring the formation of the 4-acyl cycloadducts, which are suitable intermediates for the synthesis of semi-aromatized polycyclic targets derived from polyketide type-II biosynthesis.

Full text of DOI:10.1039/c2ob25423a

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
This article is part of the
OBC 10th anniversary
themed issue
All articles in this issue will be gathered together
online at
www.rsc.org/OBC10

Dynamic ArtVicielweALritnicklesOnline
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 6003
www.rsc.org/obc
PAPER
Synthesis of isoxazoles en route to semi-aromatized polyketides:
dehydrogenation of benzonitrile oxide–para-quinone acetal cycloadducts†‡
Yoshimitsu Hashimoto, Akiomi Takada, Hiroshi Takikawa and Keisuke Suzuki*
Received 27th February 2012, Accepted 3rd April 2012
DOI: 10.1039/c2ob25423a
Avariety of highly functionalized polycyclic isoxazoles are prepared by a two-step protocol: (1)
1,3-dipolar cycloaddition of o,o′-disubstituted benzonitrile oxides to para-quinone mono-acetals, then
(2) dehydrogenation. The cycloaddition proceeds in a regioselective manner, favouring the formation of
the 4-acyl cycloadducts, which are suitable intermediates for the synthesis of semi-aromatized polycyclic
targets derived from polyketide type-II biosynthesis.
prenylbarium reagent5 and the stereospecific pinacol-type 1,2-
Introduction
shift.6 The isoxazole moiety serves not only as a 1,3-dicarbonyl
equivalent,7 but also as a directing group for the regioselective
Complex molecular architectures commonly found in biologically
active polyketide-derived natural products represent challenging
synthetic targets, which demand innovative approaches to the intro-
duction, management, and transformation of sensitive functional
arrays. Polyketide derivatives, further decorated by oxygenation of
the aromatic π-system, display novel chemical functionalities, struc-
tural complexities, and stereochemical elements that raise the level
of difficulty for the synthetic chemist.1 The cycloaddition–dehydro-
genation strategy described herein provides access to isoxazoles as
synthetic intermediates en route to polycyclic polyketide deriva-
tives of the fully-aromatized and semi-aromatized variety (Fig. 1).
Seragakinone A (1),2 a marine natural product having a
densely functionalized pentacyclic structure with an angular
carbon substituent derived from a prenyl group, provided us with
the impetus for the development of a viable strategy to construct
polycyclic systems with such architectural, functional and stereo-
chemical complexity. Two reaction processes emerged as particu-
larly valuable in this regard (Scheme 1): (1) Cyclocondensation
of benzonitrile oxide I and 1,3-diketone II gives the key isoxa-
zole intermediate III;3 (2) NHC-catalyzed benzoin reaction con-
verts to linear tricarbocycle IV with excellent enantioselectivity.4
α-Ketol IV, which corresponds to the BCD ring system in 1, is a
key platform for synthesis of angularly prenylated compound VI
via two steps including the stereoselective addition of a
1,2-shift with its ability of α-cation stabilization.
Although this model study showed a viable way for construct-
ing the BCD ring system, a serious drawback to intermediates
IV–VI was the poor functionalization at the B ring, making
further elaboration tedious, if not impossible.
Intermediate III′, a more oxidized form of III, could solve
this problem (Scheme 2, cf. Scheme 1), if para-quinone (VII)
reacts with nitrile oxide I.8 In addition, if mono-masked quinone
VII′ could be employed in this process, the functionalities in the
resulting product III′′ would be nicely discriminated and ideally
suited for further elaboration.
One could envision an indirect way for synthesizing more
functionalized isoxazole intermediates by a two-step protocol;
1,3-dipolar cycloaddition of nitrile oxides I to para-quinone
mono-acetals VII′ followed by dehydrogenation; however, the
reactivity of VII′ toward the cycloaddition to I, and oxidation of
the cycloadduct to III′′ gives one reason to pause.
Nitrile oxides tend to react with electron-deficient olefins, e.g.
α,β-unsaturated carbonyl compounds, with the interaction between
the HOMO of the dipoles and the LUMO of the dipolarophiles.
When an oxygen of nitrile oxides, which has the largest HOMO
coefficient, binds to a carbon β to a carbonyl of dipolarophiles,
a 4-acyl cycloadduct is formed as a major regioisomer.9 However,
in the reaction of acyclic enones varying degrees of regioselectivity
are observed10 to produce a mixture of 4- and 5-acylisoxazolines
(Scheme 3). Despite these concerns, several studies suggested that
reaction of cyclic enones with nitrile oxides would lead to the pre-
ferential formation of the 4-acyl cycloadducts,11 which would
make para-quinone mono-acetal VII′ a promising dipolarophile
that undergoes regioselective 1,3-dipolar cycloadditions with
nitrile oxides.12 The remaining concern was whether the presence
of substituents in the 1,3-dipoles and dipolarophiles might disturb
the regiocontrol.13
Department of Chemistry, Tokyo Institute of Technology, O-okayama,
Meguro-ku, Tokyo 152-8551, Japan. E-mail: ksuzuki@chem.titech.ac.jp;
Fax: +03 5734 2788; Tel: +03 5734 2228
†This article is part of the Organic & Biomolecular Chemistry 10th
Anniversary issue.
‡Electronic supplementary information (ESI) available. CCDC 859931
(4), 859932 (5), 859933 (9a), 859934 (8c), 859935 (8b), 859936 (8i),
874424 (8g), 874422 (2a), and 874423 (2c). For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/
c2ob25423a
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6003–6009 | 6003

View Article Online
Fig. 1 Naturally-occurring aromatic polyketides.
Scheme 3 Regioselectivity for nitrile oxides reacting with acyclic and
cyclic enones.
Scheme 4 1,3-Dipolar
cycloaddition–dehydrogenation
approach
toward functionalized isoxazoles IX.
Scheme 1 Previous work: cyclocondensation approach to isoxazole
intermediates III–VI.
Herein, we report two-step synthesis of highly functionalized
isoxazoles IX by 1,3-dipolar cycloaddition of nitrile oxide I to
para-quinone mono-acetal VII′ (step 1) followed by dehydro-
genation (step 2) (Scheme 4). This two-step operation has
already been employed in our recent synthesis of seragakinone
A,14 but we describe here its optimization, establishing a practi-
cal one-pot protocol for carrying out these two steps. Particular
emphasis is placed on its scope and limitations using various
nitrile oxides and para-quinone mono-acetals.
Results and discussion
Quinone mono-acetal 3a15 was used for the initial model study,
examining in detail the reaction with nitrile oxide 2a14 (Table 1).
Scheme 2 Quinones and masked quinones as dipolarophiles.
6004 | Org. Biomol. Chem., 2012, 10, 6003–6009
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 1,3-Dipolar cycloaddition of benzonitrile oxide 2a to p-quinone mono-acetal 3aa
Run
3a/equiv
Temp/°C
Time/h
Yieldb/%
4 : 5 : 6e
1
2
3
4
5
1.2
1.2
1.2
2.0
3.0
90
60
rt
rt
rt
2
10
8 days
4 days
2 days
65
87 : 8 : 5
88 : 7 : 5
89 : 6 : 5
90 : 5 : 5
90 : 5 : 5
70
74c,d
80c,d
84c,d
a 200 mg scale, 1.0 M. The structure of each of 4 and 5 was confirmed by X-ray crystallographic analysis.‡ b Isolated yield. c Yield of recovered 3a;
run 3: 31%, run 4: 56%, run 5: 67%. d Yield of bisisoxazoline 7; run 3: 13%, run 4: 7%, run 5: 5%. e By 1H NMR. rt: room temperature.
Scheme 5 Dehydrogenation of isoxazoline 4 by MnO2.
Fig. 2 ORTEP-Drawings of isoxazolines 4 and 5 (thermal ellipsoids
are drawn at 50% probability level).
yield increased to 70% (run 2), and the reaction at room tempera-
ture gave the cycloadducts in 74% yield, although 8 days were
necessary for the full conversion (run 3). In this case, a small
amount of a polar byproduct was noted, which was identified as
bisisoxazoline 7.17 We envisioned that this unfavorable double
cycloaddition would be suppressed by an increased molar ratio
of 3a to nitrile oxide 2a, leading to higher yield. Indeed, when
the molar ratio of 3a : 2a was made 2.0 and 3.0, the reactions
completed in shorter times (4 and 2 days, respectively), affording
the cycloadducts in 80% and 84% combined yield of 4–6,
respectively (runs 4 and 5).
Having isoxazoline 4 in hand, we examined step 2, i.e., the
dehydrogenation of isoxazoline into the corresponding isoxazole
8a (Scheme 5). To our delight, the projected reaction proceeded
smoothly by exposure of 4 to activated MnO2 (PhCl, 80 °C,
10 min),18 giving isoxazole 8a in 91% yield as the sole product.
When a mixture of 4 containing small amounts of 5 and 6
(4 : 5 : 6 = 90 : 5 : 5) was exposed to MnO2 under the same con-
ditions, the desired product 8a was obtained in 85% yield, where
a small amount of nitrile 9a was co-produced (4% yield)
(Scheme 6). To our pleasant surprise, none of the isomeric
product expectable from 5 was identified, enabling facile iso-
lation of 8a; these products 8a and 9a could be easily separated
by column chromatography (SiO2, EtOAc–hexane = 1 : 2 →
It turned out that the cycloaddition reaction proceeded in a
highly regioselective manner, providing 4-acyl isomer 4 as the
major product; upon heating of 2a and 3a in chlorobenzene at
90 °C, a mixture of three isoxazolines 4, 5, and 6 (4 : 5 : 6 =
87 : 8 : 5) was obtained in 65% yield after purification by column
chromatography (SiO2, EtOAc–hexane = 1 : 2 → 2 : 3). The
major product 4 was isolated in pure form by recrystallization
(EtOAc–hexane, colorless grains, mp 180.5–181.0 °C) or pre-
parative TLC [THF : toluene : hexane = 1 : 2 : 2 (3×)]. Preparative
TLC (Et2O) allowed isolation of small amounts of the two
regioisomers: 5-acyl isomer 5 resulted from the reversed cyclo-
addition mode to that of 4, and 5-methylisoxazoline 6 was pro-
duced by the reaction at the trisubstituted double bond in 3a.
Each structure was confirmed by extensive NMR study and/or
X-ray crystallography‡ (Fig. 2).16
For improving the yield and regioselectivity, several reaction
parameters were screened, including the solvent, the reaction
temperature, and the amount of dipolarophile 3a. While choice
of the solvent (THF, DMF, 1,4-dioxane, and toluene) gave no
improvement (not shown), higher yields were realized by lower-
ing the reaction temperature (Table 1, runs 1–3); at 60 °C, the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6003–6009 | 6005

View Article Online
much more reactive, leading to a faster reaction with 2a (run 3,
room temperature, 30 h). In this case, however, the overreaction,
that is the double cycloaddition, became serious. Pleasingly,
use of an excess amount of 3c (10 equiv) suppressed formation
of the bisisoxazoline, providing 8c in 75% yield after dehydro-
genation (run 4).
It should be noted that the C-3 substituent in 3a and 3b poses
a buttressing effect on the neighboring methoxy group at the C-4
position to enhance the steric hindrance around the reaction site.
In addition, a C-3 substituent in quinone mono-acetals prevents
the second cycloaddition, which is ascribable to the steric repul-
sion with the nitrile oxide.22
The cycloaddition reactions of nitrile oxide 2b,23 lacking a
methoxy group at the C4 position, proceeded smoothly to
quinone mono-acetals 3a–c (Table 2, runs 5–7). The reactions of
3a and 3b with 2b gave the corresponding isoxazoles 8d and 8e
in 67% and 73% yield, respectively, albeit with longer reaction
times (runs 5 and 6). Quinone mono-acetal 3c reacted rapidly
with nitrile oxide 2b even at room temperature; use of 10 equiv
of 3c efficiently suppressed the bisisoxazoline formation, afford-
ing isoxazole 8f in 77% yield (run 7).
Scheme 6 Dehydrogenation of a mixture of isoxazolines 4–6 and
possible mechanism for the formation of 9a.
Nitrile oxide 2c,3d having two methoxy substituents, proved
also amenable to this reaction (runs 8–10). It turned out that 2c
was less reactive than 2a and 2b; the reaction of 2c with quinone
mono-acetals 3a and 3b proceeded at higher temperatures
(60 and 80 °C, respectively) to afford the corresponding isoxa-
zoles 8g and 8h, while the reaction did not finish at below 40 °C
(runs 8 and 9). We rationalized that the C-5 methoxy group in 2c
posed a buttressing effect on the neighboring acetal, enhancing
the steric hindrance around the nitrile oxide moiety.24 The reac-
tion of 2c with 3c (10 equiv) occurred without event, affording
the corresponding isoxazole 8i in 83% yield after dehydrogena-
tion (run 10).
Scheme 7 One-pot synthesis of isoxazole 8a, rt: room temperature.
2 : 3). Formation of nitrile 9a could be ascribed to isomer 5, via
a
known fragmentation reaction of 5-acylisoxazolines
Conclusions
(Scheme 6, below), although the quinone A was not identifi-
ed.11a,19 This process turned out to be promoted by MnO2,
although the precise role remains unclear.20
A facile access to highly functionalized polycyclic isoxazoles has
been developed via 1,3-dipolar cycloaddition of stable benzonitrile
oxides to para-quinone mono-acetals followed by MnO2-mediated
dehydrogenation, which has paved the way to an efficient syn-
thesis of complex polycyclic semi-aromatized polyketides.
Having these results, we envisaged that the above two steps
may be carried out in one pot, which proved indeed the case;
after completion of the cycloaddition of 2a to 3a (PhCl, room
temperature, 4 days), activated MnO2 (10 equiv) was added to
the reaction mixture, and subsequent heating (80 °C, 10 min)
afforded isoxazole 8a and nitrile 9a in 67% and 6% yield,
respectively (Scheme 7). The yield of 8a was comparable to that
by the stepwise protocol (vide supra, 64% yield, 2 steps).
This one-pot protocol proved practical, working well for
nitrile oxides 2a–c and p-quinone mono-acetals 3a–c (Table 2).
The unreacted quinone mono-acetal was recovered after purifi-
cation by column chromatography. Several trends became appa-
rent: (1) the reactivity of quinone mono-acetals became slightly
lower by the presence of an electron-donating substituent at the
position β to the carbonyl, due to steric and electronic factors,
and (2) the reaction of nitrile oxides having a C-5 substituent
required higher temperatures.
Experimental section
General procedure for the one-pot synthesis of isoxazole 8a
A suspension of nitrile oxide 2a (200 mg, 0.754 mmol) and
p-quinone mono-acetal 3a (254 mg, 1.51 mmol) in chloro-
benzene (0.75 mL) was stirred at room temperature for 4 days.
To the mixture was added activated manganese(IV) oxide
(656 mg, 7.55 mmol), and stirring was continued for 10 min at
80 °C. After cooling to room temperature, the reaction mixture
was filtered through a Celite® pad (washed with MeOH). The
collected manganese residue was suspended in MeOH, which
was heated under reflux for 1 min and filtered through a Celite®
pad. The combined filtrate was condensed under reduced
pressure, and the residue was purified by flash column chromato-
graphy (silica gel, hexane–EtOAc = 2 : 1 to 3 : 2) to afford isoxa-
zole 8a (217 mg, 67%) as a colorless foam. Nitrile 9a,
Nitrile oxide 2a smoothly reacted with quinone mono-acetal
3b,21 although a higher temperature (40 °C) was needed. The
subsequent addition of MnO2 (80 °C, 10 min) gave isoxazole 8b
in 77% overall yield (run 2). Quinone mono-acetal 3c proved
6006 | Org. Biomol. Chem., 2012, 10, 6003–6009
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 One-pot synthesis of various isoxazoles via 1,3-dipolar cycloaddition followed by dehydrogenationa
Quinone mono-
acetal
Yield of
isoxazole /%
Yield of
nitrile/%
Run
1
Nitrile oxide
X/equiv
2.0
Temp./°Cb
rt
Timeb
4 days
Product
R = Me (3a)
67
ca. 6 (9a)
2
R = OMe (3b)
2.0
40
40 h
77
ca. 4 (9a)
3
R = H (3c)
2.0
10.0
rt
rt
30 h
11 h
56
75
ca. 7 (9a)
ca. 4 (9a)
4c
5
R = Me (3a)
R = OMe (3b)
R = H (3c)
2.0
rt
7 days
67
73
77
n.d.
n.d.
n.d.
6d
1.2
60
30 h
7e
10.0
rt
24 h
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6003–6009 | 6007

View Article Online
Table 2 (Contd.)
Quinone mono-
acetal
Yield of
isoxazole /%
Yield of
nitrile/%
Run
8
Nitrile oxide
X/equiv
2.0
Temp./°Cb
60
Timeb
22 h
Product
R = Me (3a)
77
5 (9c)
9
R = OMe (3b)
2.0
80
8 h
77
83
Trace (9c)
10e
R = H (3c)
10.0
rt
24 h
ca. 14 (9c)
a Step 1: unless otherwise indicated, all the reactions were performed using quinone mono-acetals (2.0 equiv to nitrile oxides) in chlorobenzene
(1.0 M) at room temperature for the indicated time. Step 2: MnO2 (10 equiv) at 80 °C for 10 min. b For step 1. c In toluene (0.4 M). d In chlorobenzene
(0.5 M) for step 2. e In chlorobenzene (0.4 M). n.d.: not detected. rt: room temperature.
contaminated with a trace impurity (11.7 mg, ca. 6%) was also
obtained. Unreacted p-quinone mono-acetal 3a was recovered
(127 mg, 50%) as yellow oil.
(No. 23000006) from Japan Society for the Promotion of
Science (JSPS), Japan. We are grateful to Ms Sachiyo Kubo,
Dr Kotaro Fujii, and Prof. Dr Hidehiro Uekusa for X-ray
analysis.
8a: Rf 0.43 (hexane–EtOAc = 1 : 1); 1H NMR (CDCl3,
300 MHz) δ 1.30 (brd, 1H, J = 12.2 Hz), 2.01–2.19 (m, 1H),
2.04 (d, 3H, J = 1.5), 3.31 (s, 3H), 3.49 (s, 3H), 3.58 (ddd, 1H,
J = 12.2, 11.3, 2.7 Hz) 3.69 (s, 3H), 3.85 (ddd, 1H, J = 11.9,
11.5, 2.4 Hz), 3.88 (s, 3H), 3.83–3.95 (m, 1H), 4.20 (dd, 1H,
J = 11.5, 5.1 Hz), 5.57 (s, 1H), 6.15 (d, 1H, J = 1.5 Hz), 6.53 (d,
1H, J = 2.4 Hz), 6.94 (d, 1H, J = 2.4 Hz); 13C NMR (CDCl3,
75 MHz) δ 15.5, 25.4, 52.0, 52.3, 55.4, 55.8, 67.0, 67.1, 95.0,
98.9, 99.2, 102.1, 106.9, 118.0, 130.4, 139.6, 152.1, 155.6,
159.1, 162.2, 172.5, 178.8; IR (ATR) 2945, 2843, 1678, 1612,
1433, 1377, 1333, 1286, 1201, 1161, 1120, 1076, 920 cm–1;
Anal. Calcd for C22H25NO8: C, 61.25; H, 5.84; N, 3.25; Found:
C, 61.18; H, 5.84; N, 3.17; mp 143 °C (decomp.) (acetone–
hexane, colorless prisms).
Notes and references
1 (a) R. Thomas, ChemBioChem, 2001, 2, 612–627; (b) C. Hertweck,
A. Luzhetskyy, Y. Rebets and A. Bechthold, Nat. Prod. Rep., 2007, 24,
162–190; (c) U. Rix, C. Fischer, L. L. Remsing and J. Rohr, Nat. Prod.
Rep., 2002, 19, 542–580; (d) J. Stauton and K. J. Weissman, Nat. Prod.
Rep., 2001, 18, 380–416; (e) B. J. Rawlings, Nat. Prod. Rep., 1999, 16,
425–484; (f) H. Zhou, Y. Li and Y. Tang, Nat. Prod. Rep., 2010, 27,
839–868.
2 (a) H. Shigemori, K. Komatsu, Y. Mikami and J. Kobayashi, Tetrahedron,
1999, 55, 14925–14930; (b) K. Komatsu, H. Shigemori, M. Shiro and
J. Kobayashi, Tetrahedron, 2000, 56, 8841–8844.
3 (a) J. W. Bode, Y. Hachisu, T. Matsuura and K. Suzuki, Tetrahedron Lett.,
2003, 44, 3555–3558; (b) J. W. Bode, Y. Hachisu, T. Matsuura and
K. Suzuki, Org. Lett., 2003, 5, 391–394; (c) T. Matsuura, J. W. Bode,
Y. Hachisu and K. Suzuki, Synlett, 2003, 1746–1748; (d) H. Takikawa,
K. Hikita and K. Suzuki, Synlett, 2007, 2252–2256.
Acknowledgements
4 (a) Y. Hachisu, J. W. Bode and K. Suzuki, J. Am. Chem. Soc., 2003, 125,
8432–8433; (b) Y. Hachisu, J. W. Bode and K. Suzuki, Adv. Synth.
Catal., 2004, 346, 1097–1100; (c) H. Takikawa, Y. Hachisu, J. W. Bode
This work was partially supported by the Global COE Program
(Chemistry) and a Grant-in-Aid for Specially Promoted Research
6008 | Org. Biomol. Chem., 2012, 10, 6003–6009
This journal is © The Royal Society of Chemistry 2012

View Article Online
and K. Suzuki, Angew. Chem., Int. Ed., 2006, 45, 3492–3494;
(d) H. Takikawa and K. Suzuki, Org. Lett., 2007, 9, 2713–2716.
5 A. Yanagisawa, S. Habaue, K. Yasue and H. Yamamoto, J. Am. Chem.
Soc., 1994, 116, 6130–6141.
6 (a) K. Suzuki, H. Takikawa, Y. Hachisu and J. W. Bode, Angew. Chem.,
Int. Ed., 2007, 46, 3252–3254; (b) H. Takikawa, K. Hikita and
K. Suzuki, Angew. Chem., Int. Ed., 2008, 47, 9887–9890.
15 M. Yu and S. J. Danishefsky, J. Am. Chem. Soc., 2008, 130, 2783–2785.
16 CCDC 859931 (4) and 859932 (5) contain the supplementary crystallo-
graphic data‡ for this paper.
17 Although the structure of 7 has not yet been clarified, mass spectroscopy
of 7 clearly suggested the 2 : 1 adduct of 2a and 3a.
18 Other oxidants [MnO2 (chemicals treated), DDQ, and chloranil] proved
less effective. For dehydrogenation of isoxazoline by MnO2, see:
(a) A. J. Fatiadi, Synthesis, 1976, 65–104; (b) A. J. Fatiadi, Synthesis,
1976, 133–167; (c) A. Barco, S. Benetti and G. P. Pollini, Synthesis,
1977, 837; (d) C. J. Easton, G. A. Heath, C. M. M. Hughes, C. K. Y. Lee,
G. P. Savage, G. W. Simpson, E. R. T. Tiekink, G. J. Vuckovic and
R. D. Webster, J. Chem. Soc., Perkin Trans. 1, 2001, 1168–1174.
19 Heating of 6 with MnO2 resulted in complex mixtures, and none of
nitrile 9a was detected.
20 Heating of a mixture of 4, 5, and 6 (4 : 5 : 6 = 90 : 5 : 5) in the absence of
MnO2 (chlorobenzene, 80 °C, 10 min) resulted in no reaction.
21 Y. Tamura, T. Yakura, J. Haruta and Y. Kita, J. Org. Chem., 1987, 52,
3927–3930.
7 For a review of isoxazole chemistry, see: (a) P. Grünanger and P. Vita-Finzi,
Isoxazole: Part 2, in The Chemistry of Heterocyclic Compounds, ed.
E. C. Taylor and P. Wipf, John Wiley & Sons, New York, 1999;
(b) P. G. Baraldi, A. Barco, S. Benetti, G. P. Pollini and D. Simoni, Synthesis,
1987, 857–869; (c) A. I. Kotyatkina, V. N. Zhabinsky and V. A. Khripach,
Russ. Chem. Rev. (Engl. Transl.), 2001, 70, 641–653; (d) B. J. Wakefield, in
Science of Synthesis: Houdenwayl Methods of Molecular Transformations,
ed. E. Schaumann, Georg Thieme, Stuttgart, 2001, vol. 11, 229–288.
8 Related studies on the 1,3-dipolar cycloaddition of nitrile oxides to qui-
nones, see: (a) S. Shiraishi, S. Ikeuchi, M. Seno and T. Asahara, Bull.
Chem. Soc. Jpn., 1978, 51, 921–925; (b) S. Shiraishi, B. S. Holla and
K. Imamura, Bull. Chem. Soc. Jpn., 1983, 56, 3457–3463;
(c) T. Hayakawa, K. Araki and S. Shiraishi, Bull. Chem. Soc. Jpn., 1984,
57, 1643–1649; (d) T. Mukawa, Y. Inoue and S. Shiraishi, Bull. Chem.
Soc. Jpn., 1999, 72, 2549–2556; (e) V. Nair, K. V. Radhakrishnan,
K. C. Sheela and N. P. Rath, Tetrahedron, 1999, 55, 14199–14210.
9 I. Fleming, Molecular Orbitals and Organic Chemical Reactions, John
Wiley & Sons, West Sussex, 2010, 334.
22 It should be noted that presence of the dimethyl acetal was essential to
prevent the undesired overreaction, while the reaction of 2a with 11 with
an ethylene acetal led to an enhanced yield of the corresponding bisisoxa-
zoline 13 (83%).
10 For
a review of 1,3-dipolar cycloaddition of nitrile oxide, see:
(a) P. Caramella and P. Grunanger, in 1,3-dipolar Cycloaddition Chemistry,
ed. A. Padwa, John Wiley & Sons, New York, 1984, vol. 1, 291–392;
(b) C. J. Easton, C. M. Hughes, G. P. Savage and G. W. Simpson, in
Advances in Heterocyclic Chemistry, ed. A. R. Katritzky, Academic Press,
New York, 1994, vol. 60, 261–327; (c) V. Jager and P. A. Colinas, in Syn-
thetic Applications of 1,3-Dipolar Cycloaddition Chemistry Toward Het-
erocycles and Natural Products, ed. A. Padwa and W. H. Pearson, John
Wiley & Sons, New York, 2002, vol. 59, 361–472; (d) L. I. Belen’KII, in
Nitrile Oxides, Nitrones, and Nitronates in Organic Synthesis: Novel Strat-
egies in Synthesis, ed. H. Feuer, John Wiley & Sons, Hoboken, 2nd edn,
2007, 1–127; (e) R. P. Litvinovskaya and V. A. Khripach, Russ. Chem. Rev.
(Engl. Transl.)., 2001, 70, 405–424.
11 (a) G. Bianchi, C. D. Micheli, R. Gandolfi, P. Grunanger, P. V. Finzi and
O. V. Pava, J. Chem. Soc., Perkin Trans. 1, 1973, 1148–1155;
(b) P. Carammella, D. Reami, M. Falzoni and P. Quadrelli, Tetrahedron,
1999, 55, 7027–7044.
12 (a) M. Schubert-Zsilavecz, D. Gusterhuber and F. Belaj, Monatsh. Chem.,
1990, 121, 555–564; (b) M. Schubert-Zsilavecz, D. Gusterhuber,
W. Likussar and G. Farber, Monatsh. Chem., 1994, 125, 735–742.
13 (a) A. Kamimura and K. Hori, Tetrahedron, 1994, 50, 7969–7980;
(b) T.-J. Lu and L.-J. Sheu, J. Chin. Chem. Soc. (Taipei), 1995, 42,
877–879.
.
23 J. W. Bode and K. Suzuki, Tetrahedron Lett., 2003, 44, 3559–3563.
24 X-Ray crystal structures of nitrile oxides 2a and 2c suggested the possible
buttressing interaction of the methoxy group on the C5 position by two
points: (1) while the ring system of the acetal in 2c is oriented to the
CNO side, that in 2a lies on the opposite side, and (2) judging from the
narrower bond angles of C2–C1–C7 and N–C7–C1 in 2c (116.9° and
168.9°, respectively) than those in 2a (118.5° and 171.2°, respectively),
the CNO group in 2c may be forced to lie closer to the C2 side.
14 A. Takada, Y. Hashimoto, H. Takikawa, K. Hikita and K. Suzuki, Angew.
Chem., Int. Ed., 2011, 50, 2297–2301; The cycloaddition of 2a to 3c
occurred in a regioselective manner, giving isoxazole 8c in 75% yield
(2 steps) after dehydrogenation by MnO2.
.
.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6003–6009 | 6009