Designed rearrangement of a spirodiepoxide

Article abstract of DOI:10.1055/s-2007-1000866

Epoxidation of a highly functionalized aryl allene gave a rearranged enone. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2007-1000866

LETTER
213
Designed Rearrangement of a Spirodiepoxide
D
esigne
d
Rearr
h
angement of
e
a
Spirodiep
n
oxide ghua Wang, Ning Shangguan, Joseph R. Cusick, Lawrence J. Williams*
Department of Chemistry & Chemical Biology, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA
E-mail: lawjw@rutgers.edu
Received 31 July 2007
enable concise strategies for the assembly of complex nat-
ural products. For example, the alkaloids shown in
Scheme 2 each house a vicinal triad (asterisks) that could,
in principle, be derived from an aryl allene by way of the
corresponding spirodiepoxide. Preliminary feasibility
studies, however, showed that DMDO3 oxidation of aryl
allene 2 did not give an isolable spirodiepoxide. Instead, a
complex mixture was obtained, including reaction prod-
ucts possibly derived from cationic intermediates, facili-
tated by the aryl substituent, and traceable to the
spirodiepoxide or the allene oxide precursor. We have re-
ported on a program that aims to understand spirodiep-
oxide reactivity and to apply this functional group to
natural product synthesis.4,5 One part focuses on the ma-
nipulation of ionization pathways potentially accessible to
spirodiepoxides, and includes as the context aryl substitut-
ed allenes. Presented here is the designed acid-induced re-
Abstract: Epoxidation of a highly functionalized aryl allene gave a
rearranged enone.
Key words: spirodiepoxide, allene, oxidation, acid-induced rear-
rangement, MCPBA
In a series of insightful studies, Crandall and co-workers
established that peracid oxidants effect the conversion of
allenes to the corresponding spirodiepoxides and that the
acid byproduct derived from the oxidant induces the con-
version of the spirodiepoxide to a range of products
(Scheme 1).1 These include mixtures of stereo- and regio-
isomers from nucleophilic opening of the spirodiepoxide
by the organic carboxylate, and rearranged products such
as oxetanone, a-hydroxy enone, and others (3–7), presum-
ably by way of carbocationic intermediates. The same
group reported that dimethyldioxirane (DMDO), in con-
trast to peracid oxidants, induces allene oxidation under
conditions that enable isolation of the spirodiepoxide.2
arrangement of
spirodiepoxide.
a
highly functionalized aryl
Key elements of the experimental design are depicted in
Scheme 3. Where relevant, a spirodiepoxide (II) should
be polarizable at the more substituted terminus under the
action of a Brønsted or Lewis acid, and the corresponding
carbocation should form upon full ionization (not shown).
Substrates with a suitably disposable group (X) adjacent
to this site would spontaneously jettison the group and
thus form a double bond (III). This process could be step-
wise or concerted. Aryl-substituted allenes were selected
as the context in which to evaluate these concepts, as this
class of allene would be of use in synthesis. Arene 2,6-di-
substitution (I, see Y) was of particular interest as a model
for a range of complex natural products. Moreover, this
substitution pattern ought to suppress spontaneous het-
erolysis of the spirodiepoxide and its precursors, a path-
way that the aryl group might otherwise promote.
a)
H
R
H
H+
•
decomposition
decomposition
R'
1
b)
TBSO
i-Bu
DMDO
•
H
2
O
O
O
R
c)
R'
R'
R'
R
R'
R
O
OH
OH
X
3
5
6
O
O
R'
R
R
Allene 25 satisfies all the above criteria (Scheme 4). The
substitution pattern matches that of the FR-mitomycino-
ids. The ether and amide groups, although potentially
problematic since they are electron donating, should in-
duce the arene to be twisted out of planarity with the al-
lene. The trimethylsilyl group should serve as the
disposable group. Moreover, the silyl cation generated in
this way could also promote spirodiepoxide opening and
lead to formation of the silyl ether (see below). As such,
the transformation would formally be a rearrangement.
X
OH
OH
4
7
Scheme 1
Conversion of allenes to spirodiepoxides and thence to
syn-a-substituted-a¢-hydroxy ketones is a single-flask
transformation of interest in chemical synthesis (8 → 9 →
10, Scheme 2).2 Reliable methods for epoxidation and
subsequent manipulation of functionalized allenes could
The twelve-step preparation of 25 began with commercial
5-nitrovanillin (15), as shown in Scheme 4. The aryl alco-
hol was converted to triflate 16. Nucleophilic aromatic
substitution (→ 17) and acetal formation (→ 18) fol-
SYNLETT 2008, No. 2, pp 0213–0216
Advanced online publication: 21.12.2007
DOI: 10.1055/s-2007-1000866; Art ID: S05507ST
© Georg Thieme Verlag Stuttgart · New York
x
x.
x
x.
2
0
0
8

214
Z. Wang et al.
LETTER
O
H
R
H
O
O
R
R'
H
H
•
R'
R
R'
OH Nu
8
9
10
OCONH2
OH
H2NOCO
O
OH
H2N
Me
OMe
*
*
*
*
O
*
*
H
NH
N
NH
N
O
O
FR-900482 (11)
mitomycin C (12)
HO
NH
OMe
O
MeO
O
HO
H
Me
O
O
N
S
AcO
H
H
H
O
Me
Me
NMe
*
*
*
*
*
*
N
Me
H
N
N
MeO
O
H
CN
O
H
OH
OH
ecteinascidin 743 (13)
cyanocycline A (14)
Scheme 2
X
X
Y
Y
Y
OH
H
O
O
•
H+
[O]
R
R
O
R
Y
Y
Y
I
II
III
Scheme 3
lowed.6 The original conditions of Heck for sp–sp2 cou- ed oxidation. The findings were discouraging: Treatment
pling proved superior to Sonogashira conditions and gave of 25 with DMDO gave several products that did not in-
enyne 20 in good yield.7 Direct coupling of alkyne 19 with clude a stable isolable spirodiepoxide; removal of the al-
17 was inferior (55% yield) to coupling with iodide 18. dehyde by reduction and subsequent conversion to the
Reduction of the nitro group to the corresponding amine benzyl ether gave an allene that, like 25, gave several
21, followed by renovation of the amino alcohol to the products but no isolable spirodiepoxide; protected vari-
O,N,N-tribenzoyl derivative and then to amide 22, proved ants of acetal 24 behaved similarly (data not shown).
convenient and efficient. Allene 24 was prepared by ep-
oxidation, protection, and then cuprate-promoted SN2¢ ad-
dition. Attempts to generate allenes from epoxide
It seemed likely that the electron-rich arene is not stable to
DMDO oxidation or, despite our design, that the spiro-
diepoxide or the allene oxide precursor derived from 25
spontaneously ionizes and goes on to give several prod-
derivatives of 20 with the nitro group in place proved
highly inefficient. Presumably, the ability of the alkyne to
ucts. DMDO oxidation of benzylic ethers and arenes is
coordinate to copper is attenuated by the electron-with-
known and could account for these observations.9 Never-
drawing nitro group and thus coordination becomes rate-
theless, in a final attempt to channel the reactivity towards
limiting. Regeneration of the aldehyde and then installa-
the enone structure, the DMDO oxidation was run in the
presence of benzoic acid. No evidence of enone was ob-
tion of the silyl protecting group gave 25.12
From the outset our interest was to use a peracid oxidant tained. An impasse had been reached with the dioxirane
to achieve the conversion of the allene to the enone. Al- oxidant. Insofar as these experiments served as precedent,
though earlier studies showed that the acid byproduct the use of peracid oxidants for this transformation seemed
from peracid oxidation of allene initiated multiple reac- much less promising than in the design stage of this study.
tion pathways, such oxidants seemed ideal for this study,
In contrast to the DMDO oxidations described above, ex-
since their use would obviate the need to add acid in a sep-
posure of 25 to MCPBA effected allene oxidation and re-
arate step.8 Still, given the unparalleled success of
arrangement to enones 26 (dr = 2.3:1, Scheme 5).10,12
DMDO, the initial focus was to induce dioxirane-mediat-
Synlett 2008, No. 2, 213–216 © Thieme Stuttgart · New York

LETTER
Designed Rearrangement of a Spirodiepoxide
215
OMe
OMe
OMe
OMe
OH
OTf
I
I
Tf2O, CH2Cl2
0 °C, 95%
HC(OMe)3, PTSA
NaI, DMSO
90 °C, 87%
H
H
H
MeO
MeOH, 100%
NO2
NO2
NO2
NO2
O
O
O
MeO
15
16
17
18
OMe
OMe
OH
OH
OH
Zn, NH4Cl, MeOH
H2O, 100%
19
Pd(OAc)2, Et3N, PPh3
THF, 71%
MeO
MeO
NO2
NH2
MeO
MeO
20
21
OMe
OMe
OH
OTBS
1) BzCl, DMAP, Et3N
CH2Cl2, 97%
O
1) MCPBA, CH2Cl2, 97%
MeO
MeO
Bz
MeO
MeO
Bz
2) TBSCl, DMAP, imidazole
CH2Cl2 92%
2) NaH, MeOH, 92%
N
N
H
H
22
23
TMS
•
TMS
MeO
MeO
H
H
•
TMSCH2MgCl, CuI
LiBr, THF, 91%
1) PTSA, acetone, 96%
MeO
MeO
H
2) TESOTf, Et3N, CH2Cl2
96%
HO
OTBS
RO
OTBS
NH
Bz
NH
Bz
O
24
25
R = TES
Scheme 4
Importantly, there was no evidence of arene oxidation. Al- to accommodate oxidation from the top face,11 and thus
though the precise stereochemical assignment of the ma- would lead predominantly to spirodiepoxide 28. Once
jor product has not been unequivocally established, the generated, the spirodiepoxide would be subject to an acid-
modest apparent selectivity suggests the following se- induced reaction pathway approximated by Scheme 3 to
quence. Epoxidation of the disubstituted terminus occurs give 26a as the major product. Indeed, it would seem that
first, an event which would be expected to take place with the protic acid byproduct of oxidation, and silyl cation
high facial selectivity and lead to the predominance of a generated subsequently, set into motion the highly effi-
single allene oxide intermediate (27, Scheme 6). This ste- cient overall rearrangement of the spirodiepoxide to the
reocenter would be destroyed upon eventual enone forma- enone product.
tion. The expected high reactivity of the allene oxide
In summary, this report discloses the first example of aryl
toward epoxidation would lead to the rapid formation of
allene spirodiepoxidation and the efficient conversion of
spirodiepoxide. The second epoxidation would be expect-
an allene to an enone. This moderately complex allene
ed to be less selective; and this stereocenter would be re-
ranks among the most densely functionalized substrates
tained in the final enone product. The neopentyl-like
evaluated to date. Weakly acidic hydroxylic solvents,5c re-
TMS-methylene substituent should block approach of the
agents that exhibit mild Lewis acid character4a or hydro-
oxidant to the bottom face (27b) in order to avoid severe
destabilizing interactions with the arene (27a).10 In con-
trast to the TMS-methylene, the aryl group should be able
gen-bonding properties,4b and now Brønsted acid reagents
have been used to promote efficient spirodiepoxide open-
TMS
MeO
MeO
OTMS
MeO
OTMS
H
MCPBA, CH2Cl2,
–78 °C to 25 °C
81%, dr = 2.3:1
•
OTBS
OTBS
H
H
O
OTES
H
O
OTES
RO
OTBS
NH
Bz
NHCOPh
NHCOPh
O
O
O
R = TES
26a
26b
25
Scheme 5
Synlett 2008, No. 2, 213–216 © Thieme Stuttgart · New York

216
Z. Wang et al.
LETTER
(2) (a) Crandall, J. K.; Batal, D. J.; Sebesta, D. P.; Lin, F. J. Org.
Chem. 1991, 56, 1153; and references cited therein. (b) See
also ref. 4 and 5.
H
R
Ar
O
R
O
[O]
Ar
•
Me
Si
H
Me
(3) DMDO oxidations were conducted using acetone as solvent.
Solutions of the oxidant were prepared by the method herein:
Murray, R. W.; Jeyaraman, R. J. Org. Chem. 1985, 50, 2847.
(4) (a) Ghosh, P.; Lotesta, S. D.; Williams, L. J. J. Am. Chem.
Soc. 2007, 129, 2438. (b) Lotesta, S. D.; Kiren, S.;
Williams, L. J. Angew. Chem. Int. Ed. 2007, 46, 7108.
(5) (a) Katukojvala, S.; Barlett, K. N.; Lotesta, S. D.; Williams,
L. J. J. Am. Chem. Soc. 2004, 126, 15348. (b) Lotesta, S.
D.; Hou, Y.; Williams, L. J. Org. Lett. 2007, 9, 869; and
references cited therein. (c) Shangguan, N.; Kiren, S.;
Williams, L. J. Org. Lett. 2007, 9, 1093.
Me
TMS
25
28
X
X
O
O
H
R
H
Me
R
Si
Me
(6) McClure, K. F.; Danishefsky, S. J. J. Am. Chem. Soc. 1993,
115, 6094.
Me
Si
Me
Me
x
Me
(7) Dieck, H. A.; Heck, F. R. J. Organomet. Chem. 1975, 93,
259.
27b
OTES
27a
BzHN
(8) (a) Tang, Y.; Marshall, J. A. J. Org. Chem. 1993, 58, 3233.
(b) Tang, Y.; Marshall, J. A. J. Org. Chem. 1994, 59, 1457.
(9) (a) Murray, R. W. Chem. Rev. 1989, 89, 1187. (b) Marples,
B. A.; Muxworthy, J. P.; Baggaley, K. H. Synlett 1992, 646.
(c) Csuk, R.; Dörr, P. Tetrahedron 1994, 50, 9983.
(10) For a closely related stereoselective allene oxidation, see ref.
5b.
CHO
Ar =
R =
OTBS
OMe
Scheme 6
(11) For example, see: Fleming, I.; Hill, J. H. M.; Parker, D.;
Waterson, D. J. Chem. Soc., Chem. Commun. 1985, 318.
(12) Selected Analytical Data:
ing. These findings contribute to a growing body of data
that demonstrate the potential of spirodiepoxides for use
in complex molecule synthesis.
Compound 25: 1H NMR (400 MHz, CCl3): d = –0.08 (s, 9
H), 0.06 (d, J = 2.1 Hz, 6 H), 0.53 (q, J = 7.9 Hz, 6 H), 0.74
(t, J = 7.9 Hz, 9 H), 0.88 (s, 9 H), 1.81 (dd, J = 1.8, 5.8 Hz,
2 H), 3.56 (dd, J = 5.7, 9.9 Hz, 1 H), 3.64 (dd, J = 5.7, 10 Hz,
1 H), 3.90 (s, 3 H), 4.18 (m, 1 H), 5.25 (td, J = 2.4, 7.8 Hz, 1
H) 7.24 (s, 1 H), 7.54 (m, 3 H), 7.89 (d, J = 7.2 Hz, 2 H), 8.68
(s, 1 H), 8.72 (br, 1 H), 10.00 (s, 1 H). IR (neat): nmax = 3403,
2721, 1953, 1700, 1581, 1259, 1001 cm–1. ESI-MS: m/z
calcd for C36H57NO5Si3Na [MNa]+: 691.0; found 690.5.
Compound 26: 1H NMR (300 MHz, CCl3): d = 0.10 (s, 3.8
H), 0.11 (s, 2.8 H), 0.42–0.54 (m, 6 H), 0.76–0.89 (m, 18 H),
3.44–3.52 (m, 1 H), 3.60–3.69 (m, 0.73 H), 3.73–3.78 (m,
1.13 H), 3.84 (s, 3 H), 4.02–4.08 (m, 0.48 H), 4.15–4.20 (m,
0.65 H), 4.73 (d, J = 2.4 Hz, 0.66 H), 4.75 (d, J = 2.4 Hz,
0.36 H), 5.88 (s, 0.68 H), 5.97 (s, 0.27 H), 6.54 (s, 0.71 H),
6.68 (s, 0.28 H), 7.45–7.58 (m, 4 H), 7.86 (d, J = 7.5 Hz, 2
H), 8.53 (s, 1 H), 8.73 (br, 0.31 H), 8.86 (br, 0.65 H), 10.01
(s, 1 H). IR (neat): nmax = 3387, 1728, 1691, 1578, 1261,
1013, 842 cm–1. ESI-MS: m/z calcd for C36H57NO7Si3Na
[MNa]+: 723.0; found: 722.3.
Acknowledgment
Generous financial support from Merck & Co. and NIH-GMS
(078145) is gratefully acknowledged. We thank Yongquan Hou for
assistance in preparing this manuscript.
References and Notes
(1) (a) Crandall, J. K.; Machleder, W. H. J. Am. Chem. Soc.
1968, 90, 7292. (b) Crandall, J. K.; Machleder, W. H.;
Thomas, M. J. J. Am. Chem. Soc. 1968, 90, 7346.
(c) Crandall, J. K.; Machleder, W. H. J. Am. Chem. Soc.
1968, 90, 7347. (d) Crandall, J. K.; Machleder, W. H.;
Sojka, S. A. J. Org. Chem. 1973, 38, 1149. (e) Crandall, J.
K.; Conover, W. W.; Komin, J. B.; Machleder, W. H. J. Org.
Chem. 1974, 39, 1723. (f) See also ref. 2.
Synlett 2008, No. 2, 213–216 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.