Enantioselective synthesis of a PKC inhibitor via catalytic C-H bond activation

Article abstract of DOI:10.1021/ol060485h

The syntheses of two biologically active molecules possessing dihydropyrroloindole cores (1 and 2) were completed using rhodium-catalyzed imine-directed C-H bond functionalization, with the second of these molecules containing a stereocenter that can be s

Full text of DOI:10.1021/ol060485h

ORGANIC
LETTERS
2006
Vol. 8, No. 8
1745-1747
Enantioselective Synthesis of a PKC
Inhibitor via Catalytic C−H Bond
Activation
Rebecca M. Wilson, Reema K. Thalji, Robert G. Bergman,* and
Jonathan A. Ellman*
Department of Chemistry and DiVision of Chemical Sciences, Lawrence Berkeley
National Laboratory, UniVersity of California, Berkeley, California 94720
rbergman@berkeley.edu; jellman@uclink.berkeley.edu
Received February 26, 2006
ABSTRACT
The syntheses of two biologically active molecules possessing dihydropyrroloindole cores (1 and 2) were completed using rhodium-catalyzed
imine-directed C H bond functionalization, with the second of these molecules containing a stereocenter that can be set with 90% ee during
−
cyclization using chiral nonracemic phosphoramidite ligands. Catalytic decarbonylation and direct indole/maleimide coupling provide efficient
access to 2.
The dihydropyrroloindole core, like its parent molecule, is
a recurring motif in drug candidates and natural products.1
For example, compounds 1 and 2 (Figure 1) have been shown
to have significant physiological effects; the former targets
the serotonin receptor (5-HT2C),2 and the latter is a protein
kinase C inhibitor selective for isozyme â.3 These activities
are potentially useful in the treatment of obesity and its
established corollary, diabetes. We recognized an opportunity
to rapidly synthesize biologically active dihydropyrrolo-
Figure 1. Biologically active dihydropyrroloindoles.
(1) For recent examples, see: (a) Bentley, J. M.; Bickerdike, M. J.;
Hebeisen, P.; Kennett, G. A.; Lightowler, S.; Mattei, P.; Mizrahi, J.; Morley,
T. J.; Plancher, J.-M.; Richter, H.; Roever, S.; Taylor, S.; Vickers, S. P.
indoles such as 1 and 2 using intramolecular catalytic C-H
Preparation of cycloalkyl fused indole derivatives and their use as 5-HT2B
and 5-HT2C receptor ligands. Int. Pat. Appl. WO 02/051844 A1, July 4,
2002. (b) Wang, Z.; Dufresne, C.; Guay, D.; Leblanc, Y. Dihydropyrrolo-
[1,2-a]indole and Tetrahydropyrido[1,2-a]indole Derivatives as Prostag-
landin D2 Receptor Antagonists. Int. Pat. Appl. WO 02/094830 A2, Nov
28, 2002. For a review on the mitomycinoids, see: (c) Danishefsky, S. J.;
Schkeryantz, J. M. Synlett 1995, 475-490.
(2) Adams, D. R.; Bentley, J. M.; Roffey, J. R. A.; Hamlyn, R. J.; Gaur,
S.; Duncton, M. A. J.; Davidson, J. E. P.; Bickerdike, M. J.; Cliffe, I. A.;
Mansell, H. L. Pyrroloindoles, Pyridoindoles, and Azopinoindoles as 5-HT2C
Agonists. US Patent 6,433,175 B1, Aug 13, 2002.
bond functionalization and our recently developed enanti-
oselective variant,4 and this paper reports the achievement
of these goals. Although there is growing interest in the
application of C-H activation in organic synthesis,5,6
examples of this important transformation in the synthesis
of biologically active molecules, particularly catalytic enan-
tioselective variants, are rare.
(3) Inaba, T.; Tanaka, M.; Sakoda, K. Disubstituted Maleimide Com-
pounds and Medicinal Utilization Thereof. Eur. Pat. Appl. 1,120,414 A1,
Aug 1, 2001.
(4) (a) Thalji, R. K.; Ahrendt, K. A.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2001, 123, 9692-9693. (b) Thalji, R. K.; Ellman, J. A.;
Bergman, R. G. J. Am. Chem. Soc. 2004, 126, 7192-7193.
10.1021/ol060485h CCC: $33.50
© 2006 American Chemical Society
Published on Web 03/15/2006

It was known from previous work in our group that
although aromatic ketimines are excellent substrates for the
rhodium-chiral phosphoramidite-mediated enantioselective
cyclization reaction,4b the corresponding aldimines are much
poorer substrates and provide products in very low yields.
After initial attempts with other phosphorus ligands yielded
unsatisfactory ee’s (<15%), we returned to the phosphor-
amidite ligands and found that by changing the metal/ligand
ratio from 1:1.5 to 1:1, we were able to observe product.
We chose to optimize the reaction using ligand L (Table 1),
which gave the best ee (73%) and the least starting material
decomposition.
Scheme 1. Synthesis of Compound 1
The synthesis of compound 1 (Scheme 1) began with the
reaction of 5-methoxyindole-3-carboxaldehyde with excess
allyl bromide to give the N-alkylated indole (3) in 97% yield.
Aldehyde 3 was then condensed with benzylamine in
quantitative yield. Subsequent treatment of the N-benzyl
imine with 5 mol % of RhCl(PPh3)3 (Wilkinson’s catalyst)
at 125 °C for 1.5 h followed by imine hydrolysis on silica
gel afforded tricyclic aldehyde 4 in 81% yield. We completed
our synthesis according to the previously reported proce-
dures;2 a Henry reaction on 4 followed by a LiAlH4 reduction
of the resulting nitroalkene afforded the final product in five
steps and in 40% overall yield.
Table 1. Directing Groups Evaluated in Asymmetric
Cyclization
In the synthesis of compound 2 (Figure 1), alkylation of
indole-3-carboxaldehyde with allyl chloride 5 provided
aldehyde 6 (Scheme 2), which was then condensed with
Scheme 2. Cyclization with Achiral Catalyst
benzylamine. We were gratified to discover that the N-benzyl
imine product, possessing an allylic methyl ether tether, is a
viable substrate for our directed C-H bond activation
reaction, cyclizing in 82% yield with 5 mol % of RhCl(PPh3)3
at 135 °C.
a Yields based on 1H NMR integration relative to 2,6-dimethoxytoluene
internal standard. Rhodium insertion into the aldimine C-H bond accounts
for a significant amount of the remaining material. b ee’s determined by
chiral HPLC after hydrolysis.
(5) For examples of the application of C-H bond activation in target-
oriented synthesis, see: (a) Harris, P. W. R.; Woodgate, P. D. J. Organomet.
Chem. 1997, 530, 211-223. (b) Johnson, J. A.; Sames, D. J. Am. Chem.
Soc. 2000, 122, 6321-6322. (c) Dangel, B. D.; Godula, K.; Youn, S. W.;
Sezen, B.; Sames, D. J. Am. Chem. Soc. 2002, 124, 11856-11857. (d)
Wehn, P. M.; Du Bois, J. J. Am. Chem. Soc. 2002, 124, 12950-12951. (e)
Ahrendt, K. A.; Bergman, R. G.; Ellman, J. A. Org. Lett. 2003, 5, 1301-
1303. (f) Pastine, S. J.; Sames, D. Org. Lett. 2003, 5, 4053-4055. (g)
O’Malley, S. J.; Tan, K. L.; Watzke, A.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2005, 127, 13496-13497.
Because we believe that reductive elimination is the rate-
limiting step in our catalytic cycle,7 and because it is well-
established that placing electron-withdrawing groups on a
metal-bound ketone can reduce the barrier to reductive elimi-
(6) For examples of the application of rhodium-carbenoid insertions into
C-H bonds in target-oriented synthesis, see: (a) Taber, D. F.; Song, Y. J.
Org. Chem. 1997, 62, 6603-6607. (b) Davies, H. M. L.; Stafford, D. G.;
Hansen, T. Org. Lett. 1999, 1, 233-236. (c) Davies, H. M. L.; Gregg, T.
M. Tetrahedron Lett. 2002, 43, 4951-4953 and references therein.
(7) (a) Kakiuchi, F.; Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda, M.;
Chatani, N.; Murai, S. Bull. Chem. Soc. Jpn. 1995, 68, 62-83. (b) Jun, C.
H.; Moon, C. W.; Hong, J. B.; Lim, S. G.; Chung, K. Y.; Kim, Y. H. Chem.
Eur. J. 2002, 8, 485-492.
1746
Org. Lett., Vol. 8, No. 8, 2006

nation from the metal center,8 we decided to modify the
N-benzyl imine in an effort to increase the efficiency of the
cyclization.
give both the highest ee and the highest yield. Imine 8 was
then cyclized using 10% mol % of [RhCl(coe)2]2 and 20%
mol % of the enantiomer of ligand L (ligand L′) and
hydrolyzed with 10% acetic acid in THF to afford the final
product in 61% isolated yield and 90% ee.
As expected, placing a trifluoromethyl group at the para
position increased the yield significantly, from 31% to 44%
at 105 °C (Table 1, entry 3), while a methoxy group at the
same position caused the yield to drop slightly to 28% (entry
1). Placing trifluoromethyl groups at both meta positions
caused another significant increase in yield, this time giving
the cyclized product in 57% yield with 86% ee (entry 4).
We attribute this yield increase partly to the more electron-
withdrawing nature of the substituent and partly to the fact
that, unlike most of the other substrates, the bis(trifluoro-
methyl)benzyl imine can be readily purified by recrystalli-
zation from pentane. Decreasing the reaction temperature to
90 °C afforded the desired product in 65% yield and 90%
ee. The methylene moiety of the benzyl group appears to be
critical in the transfer of chirality from catalyst to product;
although the imine derived from aniline gives a much higher
yield than does the unsubstituted N-benzyl imine, the ee
drops precipitously (entry 5).
To determine the absolute stereochemistry of the cyclized
product, compound 7 was condensed with 4-bromo-2-
nitrophenylhydrazine, and X-ray quality crystals of the
resulting hydrazone (12, see the Supporting Information)
were grown. This compound was shown by anomalous
dispersion to have an (S)-configuration as a result of the (R)-
BINOL-based ligand used in the cyclization; this is consistent
with previous work done on this type of system.4b
To perform the catalytic decarbonylation, literature condi-
tions for indole-2-carboxaldehydes were applied to our sub-
strate (S-7),9 giving the desired decarbonylated product (9)
in 86% yield. Although initial attempts at 3-lithiation of 9
proved unsuccessful, the direct substitution of indole 9 into
maleimide 10, a transformation which has not been reported
for an N-alkylated indole, gave 11 in good yield in THF at
85 °C over 5.5 days. To complete the synthesis, 11 was cross-
coupled with aniline10 and subsequently deprotected with
methanesulfonic acid,11 providing compound 2 in eight linear
steps.
As a result of the directing group studies, we chose to use
bis(trifluoromethyl)benzyl imine 8 (Scheme 3) in our syn-
In conclusion, we have demonstrated the use of catalytic
directed C-H bond activation in the synthesis of two
biologically active tricyclic indoles, the second of which
contains a stereocenter that can be set with high ee (90%)
by the use of enantioselective catalysis. The C-H activation
methods described in this publication should be applicable
to the efficient syntheses of a variety of analogues of 1 and
2 as well as to the synthesis of other biologically active
dihydropyrroloindoles.
Scheme 3. Enantioselective Cyclization and Synthesis of
Compound 2
Acknowledgment. This work was supported by the NIH
GM069559 (to J.A.E.) and the Director and Office of Energy
Research, Office of Basic Energy Sciences, Chemical Sci-
ences Division, U.S. Department of Energy, under Contract
No. DE-AC02-05CH11231 (to R.G.B.). We thank Dr. Allen
Oliver and Dr. Fred Hollander of the UC Berkeley
CHEXRAY facility for carrying out the X-ray diffraction
studies.
Supporting Information Available: Complete experi-
mental details and spectral data for all compounds described
in the paper. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL060485H
thesis; this molecule was made from 6 and the commercially
available 3,5-bis(trifluoromethyl)benzylamine in 85% yield
after recrystallization. Although other phosphoramidite ligands
were tested with this new substrate, ligand L was found to
(9) Meyer, M. D.; Kruse, L. I. J. Org. Chem. 1984, 49, 3195-3199.
(10) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1144-1157.
Although Pd(OAc)2 proved to be an excellent catalyst precursor for the
cross-coupling of a maleimide subtrate, Pd2(dba)3 gave 0% conversion under
identical conditions.
(11) Shen, L.; Prouty, C.; Conway, B. R.; Westover, L.; Xu, J. Z.; Look,
R. A.; Chen, X.; Beavers, M. P.; Roberts, J.; Murray, W. V.; Demarest, K.
T.; Kuo, G. H. Bioorg. Med. Chem. 2004, 12, 1239-1255.
(8) Lenges, C. P.; Brookhart, M. J. Am. Chem. Soc. 1999, 121, 6616-
6623.
Org. Lett., Vol. 8, No. 8, 2006
1747