then be performed through deprotonation of the dienyl
system, followed by transmetalation with a suitable chiral
metal complex (e.g., ML*dTi(OR*)4,...) and reaction of the
resulting pentadienyl-metal complex with various electro-
philes.5b-d
Spirocyclic oxindoles were prepared using a methodology
based on a SmI2-mediated aryl cyclization process. Tanaka
et al.8 recently reported that SmI2-HMPA was able to
mediate the radical cyclization of iodoamides 2 to afford the
spirocyclic skeleton of 3, provided that an ortho substituent
was present on the arene (Scheme 2). 2 (R ) R′ ) H) was
of the base. n-BuLi and s-BuLi were shown to afford a
complex mixture of products, while LiHMDS, NaH, and
t-BuOLi led to recovered starting material in essentially
quantitative yield. In contrast, metalation with t-BuOK
followed by addition of MeI led to a mixture of the C-4
methylated spiro oxindole (20%) and the 2,4-isomerized spiro
oxindole (53%). More attractive results were obtained when
LDA was used as a base. When 5a was treated with LDA at
-78 °C, no reaction took place. However, upon warming to
-40 °C, the reaction mixture turned rapidly deep purple, a
color which rapidly vanished upon addition of MeI. To our
surprise, a rearranged product 6a, possessing a new 1,4-
cyclohexadienyl ring substituted at C-3, was formed without
a trace of the expected methylated spiro oxindole (Table 1).
Scheme 2. Preparation of Spirocyclic Oxindoles 5a,b
Table 1. LDA-Mediated Rearrangement of Spiro Oxindoles
5a,b, Followed By Alkylation
entry spiro oxindole
R′
RX
MeI
AllylBr Allyl
MOMCl MeOCH2
R
diene yielda
1
2
3
4
5a
5a
5a
5b
H
H
H
Me
6a
6b
6c
6d
67
48
30
42
OMe MeI
Me
a Isolated yield of 6a-d after chromatography.
effectively shown to provide the desired oxindole 5a, along
with large amounts of the unseparable phenanthridinone 4.
In contrast, the expected spiro compounds 3a,b were
obtained in reasonable to good yields from amides 2a,b
having a bulky silyloxy group (OTBDMS) in the ortho
position. Silyloxycyclohexadienes 3a,b were then trans-
formed into the desired cyclohexadienes 5a,b following a
two-step procedure, including the conversion of 3a,b into
the corresponding triflates (not shown), using a fluoride
source and PhN(Tf)2,9 followed by a palladium-mediated
hydrogenation of the latter with formic acid.10
Varying the nature of the electrophile, a range of alkylated
products 6a-d was obtained in moderate to good isolated
yields but in all cases as a single regioisomer (C3). Reactions
were generally clean (crude yield >90%), the isolated yield
not reflecting the efficiency of the process due to the
sensitivity of dienes 6a-d to rearomatization. The structure
of regioisomers 6a-d was assigned based on the X-ray
structure determination of a derivative of 6a (vide infra).
Metalation and methylation of spiro oxindole 5b also led to
rearranged product 6d after methylation. Other electrophiles
such as aldehydes also reacted, providing alcohols 7a-d with
complete regiocontrol and in most cases with high diaste-
reocontrol (Table 2). Surprisingly and in contrast with
alkylation above, reaction with aldehydes took place at C-5.11
The stereochemistry of 7a-d was assigned based on X-ray
diffraction studies performed on alcohol 7a.12 Reaction of
5a with formaldehyde invariably led to the formation of the
phenanthridinone 4 and no trace of the desired alcohol.
Interestingly, cinnamaldehyde reacted with rearranged 5a
in a 1,4-fashion to give the aldehyde 8 with complete regio-
With spirocyclic oxindoles 5a,b in hand, we then turned
our attention toward the metalation step, varying the nature
(5) (a) Abd Rahman, N.; Landais, Y. Curr. Org. Chem. 2002, 6, 1369–
1395. (b) Studer, A.; Schleth, F. Synlett 2005, 3033–3041. (c) Maji, M. S.;
Frohlich, R.; Studer, A. Org. Lett. 2008, 10, 1847–1850. (d) Umeda, R.;
Studer, A. Org. Lett. 2007, 9, 2175–2178. (e) Butters, M.; Elliott, M. C.;
Hill-Cousins, J.; Paine, J. S.; Walker, J. K. E. Org. Lett. 2007, 9, 792–803,
and references cited therein.
(6) Epoxidation of II (R ) Me, R′ ) H) using CF3CO3H was shown to
lead exclusively to the expoxidation syn relative to the CdO group, likely
as a result of the amide functional group directing effect: Rousseau, G.;
Robert, F.; Landais, Y. unpublished results.
(7) Clayden, J. Organolithiums: Selectivity for Synthesis. Tetrahedon
Organic Chemistry Series; Elsevier Science: Oxford, 2002; Vol. 23.
(8) (a) Ohno, H.; Iwasaki, H.; Eguchi, T.; Tanaka, T. Chem. Commun.
2004, 2228–2229. (b) Iwasaki, H.; Eguchi, T.; Tsutsui, N.; Ohno, H.;
Tanaka, T. J. Org. Chem. 2008, 73, 7145–7152.
(11) (a) Snider, B. B.; Phillips, G. B.; Cordova, R. J. Org. Chem. 1983,
48, 3003–3010. (b) Ballester, P.; Costa, A.; Garcia-Raso, A.; Gomez-
Solivellas, A.; Mestres, R. Tetrahedron Lett. 1985, 26, 3625–3628.
(12) The relative configuration of alcohol 7a results from an ul-addition,
involving the Si-face of the enolate and the Re-face of the aldehyde, through
a transition state in which the aldehyde carbonyl group is likely coordinated
to the lithium cation. Seebach, D.; Prelog, V. Angew. Chem., Int. Ed. Engl.
1982, 21, 654–660.
(9) Mi, Y.; Schreiber, V.; Corey, E. J. J. Am. Chem. Soc. 2002, 124,
11290–11291.
(10) Mori, R.; Nakanishi, M.; Kajishima, D.; Sato, Y. J. Am. Chem.
Soc. 2003, 125, 9801–9807.
4442
Org. Lett., Vol. 10, No. 20, 2008