low conversions of substrate (Table 1, entries 3-5). Bulky
diphosphine ligand XANTPHOS usually used for n-selective
hydroformylations of terminal olefins yielded in 1/5.2 ratio
of 5a/6a but in poor yield of 37%. Phosphite ligand P(OPh)3
and bulky diphosphite BIPHEPHOS gave good yields but
had low influence on regioselectivity of the reaction (Table
1, entries 7 and 8). Because PPh3 appeared to be the best
compromise between yield and selectivity of hydroformy-
lation step, this ligand was chosen for further investigations.
Surprisingly, in all cases, 1-substituted carboline 5a was
isolated as racemate while 3-substituted 6a retained enan-
tiopurity of substrate. Racemization of the 1-substituted
carbolines was also observed by Danishevsky et al. during
their investigation on this type of rearrangement. The authors
proposed two possible rearrangement pathways of the
intermediate enantiopure 3,3-spiroindoleninium cations 7
(Scheme 3).9f Wagner-Meerwein like 1,2 shift proceeds in
and occurs presumably via C1-Nb bond scission.2b To test
whether racemization in our case is promoted by the acid
used in indolization step, stepwise procedure involving
thermally induced indolization9c was attempted starting from
4a, but no products were obtained. Lewis acid catalyzed
indolization with the same substrate gave the desired products
5a and 6a but again 1-substituted THBC was isolated as a
racemate. Stereochemical outcome is also determined by the
rearrangement pathway. To resolve whether retro Mannich
reaction or 1,2 shift are occurring, a crossover experiment
with a mixture of the acyclic, symmetric allylic amines was
performed.12 This experiment indicates that the rearrange-
ment occurs in an intramolecular fashion. This is in ac-
cordance with previous studies, which also showed that the
rearrangement of the 3,3-spiroindoleninium cations is an
intramolecular proces.9a This observation therefore excludes
the retro Mannich pathway and indicates that racemization
is most probably a post rearrangment event, in this case
caused by the acid used for indolization. The stereocenter
of 3-substituted carboline product is not influenced by any
of the possible reaction pathways; therefore, it stays intact.
We next turned our attention to the scope of the reaction
(Table 2). Several substrates with various alkyl or aromatic
Scheme 3. Possible Rearrangement Pathways
Table 2. Syntheses of THBCs from Substrates 4a-g, a′
olefin
yield
5 + 6c
85 (a)
ee, %e
entry
R
PG
Ts
R1
Ha
5: 6d
5
6
a suprafacial fashion,10 and leads to preservation of the
stereochemical information. A second mechanistic pathway
involves retro Mannich reaction of 7 to give the achiral, ring
opened intermediate 8, followed by subsequent cyclization
directly yielding 9rac. However, attack of the iminium ion
8 can also occur at position 311 of indole core, giving back
7. This implies that an equilibrium between 7 and 8 might
exist, leading to racemization of the spiroindoleninium cation
7 as a consequence. In addition, another factor that has to
be taken into consideration is the acid-promoted racemization
of 1-substituted THBCs. This is a well-documented process
1
2
3
4
5
6
7
8
9
Ph
Ph
1:2.7
0
0
0
0
0
0
0
0
/
97
98
96
95
ndg
ndg
94
93
95
95
Eoc Ha
63 (a′) 1:2.1
p-MeO-Ph Ts
Ha
Ha
Ha
Ha
Ha
Ha
76 (b)
71 (c)
74 (d)
58 (e)
65 (f)
69 (g)
1:5.5
1:5.2
1:2.4
1:1.4
1:3.3
1:3.6
o-MeO-Ph
p-Cl-Ph
p-CF3-Ph
Et
Ts
Ts
Ts
Ts
Ts
Ts
Ts
nPr
Ph
Ph
o-Meb 45 (h)
/
f
10
p-Clb
41 (i)
1:4.4
0
a Conditions: 1 equiv 4, 1 equiv phenyl hydrazine, 1 mol % Rh(acac)-
(CO)2, 5 mol % PPh3, 50/10 bar CO/H2, THF, 100 °C, 3 d then 4 w %
H2SO4, THF, 80 °C. b Tandem reaction run with 1 equiv of PTSA and 1
equiv of benzhydrylidene protected phenyl hydrazine. c Isolated yield after
column chromatography. d Determined by isolation. e Determined by HPLC
on chiral column. f Only 3-substituted product isolated. g Not determined.
(9) For some previous work considering this type of rearrangement, see:
(a) Jackson, A. H.; Smith, A. E. Tetrahedron 1968, 24, 403. (b) Jackson,
A. H.; Smith, P. Tetrahedron 1968, 24, 222. (c) Benito, Y.; Temporano,
F.; Rodriguez, J. G. J. Heterocyclic Chem, 1985, 22, 1207. (d) Benito, Y.;
Canoira, L.; Martinez-Lopez, N.; Temporano, F.; Rodriguez, J. G. J. Het-
erocyclic Chem, 1987, 24, 623. (e) Lynch, P. P.; Jackson, A. H. J. Chem.
Soc., Perkin Trans. 1 1987, 1215. (f) Li, C.; Chan, K.; Heimamm, A. C.;
Danishefsky, S. J. Angew. Chem., Int. Ed. 2007, 46, 1444–7. (g) Liu, K. G.;
Robichaud, A. J.; Lo, J. R.; Mattes, J. F.; Cai, Y. Org. Lett. 2006, 8, 5769–
71.
groups at the pyrrole ring were tested using the optimized
conditions in the presence of unsubstituted phenyl hydrazines
(10) Starling, S. M.; Vonwiller, S. C.; Reek, J. N. H. J. Org. Chem.
1998, 63, 2262.
(12) See Supporting Information for details on crossover experiment.
(13) (a) Wagaw, S.; Yang, B. H.; Buchwald, S. L. J. Am. Chem. Soc.
1998, 120, 6621–6622. (b) Wagaw, S.; Yang, B. H.; Buchwald, S. L. J. Am.
Chem. Soc. 1998, 121, 10251–10263.
(11) (a) van Maarseveen, J. H.; Scheeren, H. W.; Kruse, C. G.
Tetrahedron 1993, 49, 2325. (b) He, F.; Bo, Y.; Altom, J. D.; Corey, E. J.
J. Am. Chem. Soc. 1999, 121, 6771. (c) Amat, M.; Santos, M. M. M.;
Gomez, A. M.; Jokic, D.; Molins, E.; Bosch, J. Org. Lett. 2007, 9, 2907–
10.
(14) (a) Schmidt, A. M.; Eilbracht, P. Org. Biomol. Chem. 2005, 3, 124–
134. (b) Schmidt, A. M.; Eilbracht, P. J. Org. Chem. 2005, 70, 5528–5535.
Org. Lett., Vol. 10, No. 16, 2008
3435