New spirocyclic oxindole synthesis based on a hetero claisen rearrangement

Article abstract of DOI:10.1021/ol0491888

(Equation Presented) A new method for preparing spirocyclic oxindoles is presented. Featuring a [3,3]-sigmatropic enolate rearrangement, the three-step process converts carboxylic acid starting materials to oxidnole products in overall yields of 52-76%. The enolate rearrangement step occurs at -78 °C and provides easy access to oxindole products that have previously been difficult to prepare.

Full text of DOI:10.1021/ol0491888

ORGANIC
LETTERS
2004
Vol. 6, No. 14
2425-2428
New Spirocyclic Oxindole Synthesis
Based on a Hetero Claisen
Rearrangement
Zhan Mao and Steven W. Baldwin*
Paul M. Gross Chemical Laboratory, Duke UniVersity,
Durham, North Carolina 27708-0346
steVen.baldwin@duke.edu
Received May 4, 2004
ABSTRACT
A new method for preparing spirocyclic oxindoles is presented. Featuring a [3,3]-sigmatropic enolate rearrangement, the three-step process
converts carboxylic acid starting materials to oxidnole products in overall yields of 52−76%. The enolate rearrangement step occurs at −78
°C and provides easy access to oxindole products that have previously been difficult to prepare.
Incorporating the oxindole substructure into relevant alkaloids
can be accomplished by several different general strategies.
In some cases, particularly when an existing indole alkaloid
is being converted to a related oxindole alkaloid in a
semisynthetic process, it is sometimes possible to effect an
oxidative ring contraction as was done in the conversion of
the indole alkaloid gardnerine to the oxindole alkaloid
gelsemicine.1 In true total syntheses, some investigators have
chosen to incorporate elements of the oxindole moiety into
early-stage synthetic intermediates,2 while other investigators
have relegated the bulk of the oxindole construction to a
late stage in their synthetic program.
formation has been accomplished only once in a conceptually
elegant but relatively inefficient manner.3 Fleming has
reported several other approaches to this problem, although
their direct application to gelsemine or other oxindole alka-
loids has not appeared.4
While considering alternative strategies to solve this prob-
lem, we were attracted by the possibility that a [3,3] sig-
matropic rearrangement of an intermediate such as 3 (M )
metal, SiMe3) would produce a substituted aniline derivative
such as 4 suitable for cyclization to spirocyclic product 5.
Formally related to the Fischer indole synthesis5 and the
Brunner oxindole synthesis,6 this 3-aza-4-oxa-[3,3] sigma-
(1) (a) Kitajima, M.; Takayama, H.; Sakai, S. J. Chem. Soc., Perkin
Trans. 1 1994, 1573-1578. (b) Takayama, H.; Masubuchi, K.; Kitajima,
M.; Aimi, N.; Sakai, S. Tetrahedron 1989, 45, 1327-1336.
(2) See, for example, (a) Fukyama, T.; Liu, G. J. Am. Chem. Soc. 1996,
118, 7426-7427. (b) Ng, F. W.; Lin, H.; Danishefsky, S. J. J. Am. Chem.
Soc. 2002, 124, 9812-9824.
(3) (a) Sheikh, Z.; Steel, R.; Tasker, A. S.; Johnson, A. P. J. Chem. Soc.,
Chem. Commun. 1994, 763-764. (b) Dutton, J. K.; Steel, R. W.; Tasker,
A. S.; Popsavin, V.; Johnson, A. P. J. Chem. Soc., Chem. Commun. 1994,
765-766.
(4) (a) Fleming, I.; Loreto, M. A.; Michael, J. P.; Wallace, I. H. M.
Tetrahedron Lett. 1982, 23, 2053-2056. (b) Clarke, C.; Fleming, I.;
Fortunak, J. M. D.; Gallagher, P. T.; Honan, M. C.; Nuebling, C. O.;
Raithby, P. R.; Wolff, J. J. Tetrahedron 1988, 44, 3931-3944.
(5) (a) Fischer, E.; Jourdan, F. Ber. 1883, 16, 2241-2245. (b) Robinson,
B. Chem. ReV. 1969, 69, 227-250.
With regard to the latter, such an approach can often
require the net conversion of a cyclic ketone to a spirocyclic
oxindole as illustrated by the conversion of tetracyclic ketone
1 to the alkaloid gelsemine (2). In fact, this specific trans-
10.1021/ol0491888 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/11/2004

tropic process7 is expected to be particularly facile because
of anion acceleration when M ) metal and the weak N-O
bond.8
Scheme 1
In fact, similar rearrangements have been previously
reported. For instance, in 1977, Coates described the suc-
cessful rearrangement of compound 6 at 110 °C (presumably
through the enol tautomer),9 and in 1984, Blechert reported
a similar room-temperature rearrangement of stabilized anion
7.10 More recently, Endo has published the results of similar
processes of which substrate 8 is typical.11 Similarly,
Prabhakar has reported the rearrangement of substrates such
as 9.12 In the latter case, the authors suggested that the
rearrangement worked best when the enolate X group could
stablize an anion (Ph, SPh, double bond). None of the
previous investigators, however, developed the reaction as
a general route to spirocyclic oxindoles.
carbamate 12. Treatment of 12 with potassium hexameth-
yldisilazide (KHMDS; 1.1 equiv) in THF at -78 °C fol-
lowed by warming to room temperature afforded an 82%
yield of a new product that was assigned the desired
rearranged structure 13 on the basis of its spectral charac-
teristics. Treatment of 13 with sodium acetate/acetic anhy-
dride at 60-70 °C for 3 h then gave protected spirocyclic
oxindole 14. Carbonyl absorptions at 1794, 1762, and 1731
cm-1 are consistent with the assigned structure.14 A one-
proton doublet at δ 7.87 (J ) 8.0 Hz) is assigned to the
proton at position 7 of the protected spirocycle. Confirmation
of the structure was obtained by removing the carbamate
protecting group to afford lactam 15, whose analytical
characteristics were identical to those previously reported
for this material.4,15
With the above as background, we have prepared a series
of N,O-diacylated phenyl hydroxylamine derivatives de-
signed to determine the feasibility of using this 3-aza-4-oxa
[3,3] sigmatropic rearrangement to prepare spirocyclic ox-
indole derivatives. The overall transformation is illustrated
for the conversion of cyclohexane carboxylic acid (11) to
protected spiro oxindole 14 (Scheme 1).
The [3,3] rearrangement takes place at low temperature.
Thus, when the potassium enolate derived from 12 was
allowed to stir at -78 °C for 1.5 h followed by quenching
with cold methanol at that temperature, the only ma-
1
terial observed in the H NMR of the crude product mix-
Thus, DCC condensation of hydroxamic acid 1013 with
cyclohexane carboxylic acid (11) afforded the O-acylated
ture was rearranged 13. Careful inspection of the crude
spectra revealed no remaining starting material. Without
purification of the two intermediate compounds, the three-
step conversion of acid 11 to 14 was accomplished in 76%
overall yield.
In a similar fashion, adamantane carboxylic acid (16) was
converted in three steps to spirocyclic oxindole 17 (70%
yield, unoptimized) and cycloheptane carboxylic acid (19)
was smoothly transformed into the corresponding oxindole
20 (61% yield, unoptimized) (Scheme 2). In the case of the
adamantane derivative, removal of the methyl carbamate
(NaCN/DMSO; 160 °C; 2 h)14b afforded the previously
reported lactam 18.4
(6) Brunner, K. Monatsh. Chem. 1896, 17, 479-490.
(7) For a review of similar hetero [3,3] rearrangements, see: Blechert,
S. Synthesis 1989, 71-82.
(8) The bond dissociation energy of the N-O bond has been estimated
to be approximately 43 kcal/mol (180 kJ/mol). Lowry, T. H.; Richardson,
K. S. Mechanism and Theory in Organic Chemistry, 3rd ed.; Harper and
Row: New York, 1987; p 162.
(9) Coates, R. M.; Said, I. M. J. Am. Chem. Soc. 1977, 99, 2355-
2357.
(10) Blechert, S. Tetrahedron Lett. 1984, 25, 1547-1550.
(11) (a) Endo, Y.; Hizatate, S.; Shudo, K. Tetrahedron Lett. 1991, 32,
2803-2806. (b) Uchida, T.; Endo, Y.; Hizatate, S.; Shudo, K. Chem. Pharm
Bull. 1994, 42, 419. (c) Endo, Y.; Uchida, T.; Hizatate, S.; Shudo, K.
Synthesis 1994, 1096-1105.
(12) (a) Almeida, P. S.; Prabhakar, S.; Lobo, A. M.; Marcelo-Curto, M.
J. Tetrahedron Lett. 1991, 32, 2671-2674. (b) Lobo, A. M.; Prabhakar, S.
Pure Appl. Chem. 1997, 69, 547-552. (c) Santos, P. F.; Almeida, P. S.;
Lobo, A. M.; Prabhakar, S. Heterocycles 2001, 55, 1029-1043.
(13) Aresta, M.; Berloco, C.; Quaranta, E. Tetrahedron 1995, 51, 8073-
8088.
(14) (a) Rajeswaran, W. G.; Cohen, L. A. Tetrahedron 1998, 54, 11375-
11380. (b) Morales-Rios, M. S.; Bucio, M. A.; Joseph-Nathan, P.
Tetrahedron 1996, 52, 5339-5348.
(15) Moore, R. F.; Plant, S. G. P. J. Chem. Soc. 1951, 3475-3478.
2426
Org. Lett., Vol. 6, No. 14, 2004

major isomer, 25, was tentatively assigned to the isomer with
an axial aryl substituent on the basis of the comparison of
its H NMR spectrum with that of adamantane derivative
Scheme 2
1
17.
1
A consistent feature of the H NMR spectra of all of the
carbamate-protected spirocycles reported herein is a one-
proton doublet in the δ 7.8-8.0 range assignable to the H-7
proton of the aromatic oxindole ring. In adamantane deriva-
tive 17, this doublet occurrs at δ 7.91 and is accompanied
by a second one-proton doublet at δ 7.79 assignable to the
H-4 proton. Situated proximate to and bifurcating two
adamantane 1,3-diaxial protons, the H-4 proton of 17 is
apparently significantly deshielded compared to those of 14
and 20, which occur at ca. δ 7.35.
In the tert-butylcyclohexane series, the corresponding H-4
proton of the major isomer 25 is similarly situated with
respect to two 1,3-diaxial cyclohexane protons, and as
expected, the doublet for this proton occurs downfield at δ
7.52. The corresponding H-4 proton in minor isomer 26 is
seen at δ 7.20, in accord with the H-4 doublets at δ 7.31
and 7.39 for oxindoles 14 and 20, respectively.
Scheme 3
Norbornane derivatives 22 and 23 also underwent effi-
cient three-step spiroannelation processes to produce two
pairs of diastereomers, 27/28 and 29/30, respectively. In the
case of norbornane carboxylic acid 22, the major oxindole
isomer 27 is the result of a [3,3] sigmatropic process
occurring from the less hindered top face of the norbornane
ring, while the major rearrangement isomer 30 from nor-
bornene carboxylic acid 23 is the result of the [3,3] process
occurring from the less hindered bottom face. The ratios of
the diastereomers in these two cases were determined by
converting each diastereomeric pair to the known unpro-
tected spirooxindoles 33 and 34 that had been previously
reported individually and unambiguously by Fleming.4
1
Integration of H NMR signals of the 7-syn protons in the
two isomers (doublets, J ) 10 Hz) provided the ratios
reported above.
Exposure of the more sterically demanding camphor
carboxylic acid 24 (mixture of endo and exo isomers) to the
same sequence of reactions afforded the spiroannelated
oxindoles 31 and 32 in 52% yield for the three steps. As
with the norbornane isomers discussed above, the identities
and relative amounts of the two product stereoisomers were
determined by removal of the carbamate protecting group
followed by careful chromatographic separation and weigh-
ing of the individual parent oxindoles 35 and 36.
The stereochemistry of the spiroannelation process was
next investigated. Carboxylic acids 21-24 (mixtures of
isomers) were exposed to the three-step process described
above for cyclohexane derivative 11 (Scheme 3). In all cases,
the yields of the protected spirocyclic oxindoles were 50%
or greater. tert-Butylcyclohexane derivative 21 was converted
to a 2.0:1 mixture of spirocyclic isomers 25 and 26, the ratio
determined by the weights of the individual isomers isolated
after flash chromatography.16 The stereochemistry of the
(16) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-
2925.
Org. Lett., Vol. 6, No. 14, 2004
2427

phenyl hydrazides to 205-210 °C led to improved yields in
some cases.17 Under these conditions, the spirooxindole 15
from cyclohexane carboxylic acid phenyl hydrazide 37 was
formed in 51% yield.
Although oxindoles 35 and 36 have been previously
reported,17 neither had been isolated in pure form. Moreover,
the previous structure assignments based on 1H NMR
experiments were ambiguous. Accordingly, each pure isomer
was analyzed in an attempt to arrive at a secure structural
assignment. Of particular interest was the observation that
irradiation of the δ 1.45 methyl singlet (7-syn) in the minor
isomer 35 led to significant NOE enhancement of the indole
H-4 doublet at δ 7.26 (J ) 7.6 Hz). This result is consistent
with structure 35, the result of [3,3] rearrangement from the
more hindered exo face of the camphor molecule. No such
NOE enhancements were observed on irradiation of the
remaining five methyl singlets in major and minor isomers
35 and 36. The major isomer 36, therefore, arises as expected
from a rearrangement process that occurs from the less
hindered endo face of the camphor molecule.
It is interesting to compare the formation of spirocyclic
oxindoles by the [3,3] sigmatropic rearrangement process
described herein with the related Brunner oxindole synthesis.
First reported in late nineteenth century,6 the Brunner
oxindole synthesis involves the treatment of acyl phenyl
hydrazides with strong bases such as calcium oxide, calcium
hydride, and sodium amide, among others, followed by
heating to temperatures of 200 °C or higher.18 For instance,
in an optimized experimental procedure, the phenyl hydrazide
of propanoic acid was mixed with calcium hydride and then
heated to 230 °C to give a 41-44% yield of 3-methylox-
indole.19 In a more recent study of the reaction, it was
reported that heating the preformed lithium salts (nBuLi) of
By contrast, the [3,3] sigmatropic rearrangement reactions
described here occur at -78 °C, nearly 300 °C lower than
those of the Brunner oxindole reaction. In the synthesis of
complex targets such as gelsemine (1), the more gentle
reaction conditions are clearly preferred.
In summary, the rearrangements of the potassium enolates
of N-protected O-acyl phenyl hydroxylamines occurs under
mild conditions and with good efficiencies. The products are
easily cyclized to spirocyclic oxindoles. The method is
suitable for application to complex target molecules.
Acknowledgment. We acknowledge NSF (CHE 95
22580) for assistance in the purchase of the 400 MHz NMR
instrument used in this work.
Supporting Information Available: Typical experimen-
tal procedures and structural characterization data for com-
1
pounds 12-14, 17, 20, 25-32, and 35-36 and H NMR
spectra for other compounds related to 12 and 27-32. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0491888
(17) Wolff, J.; Taddei, M. Tetrahedron 1986, 42, 4267-4272.
(18) Sundberg, R. J. The Chemistry of Indoles; Academic: New York,
1970; p 362.
(19) Endler, A. S.; Beeker, E. I. Org. Synth. 1957, 37, 60-62.
2428
Org. Lett., Vol. 6, No. 14, 2004