Multicatalytic asymmetric synthesis of complex tetrahydrocarbazoles via a Diels-Alder/benzoin reaction sequence

Article abstract of DOI:10.1021/ol300192p

Expanding upon the recently developed aminocatalytic asymmetric indole-2,3-quinodimethane strategy, a straightforward synthesis of structurally and stereochemically complex tetrahydrocarbazoles has been devised. The chemistry's complexity-generating power was further harnessed by designing a multicatalytic, one-pot Diels-Alder/benzoin reaction sequence to stereoselectively access trans-fused tetracyclic indole-based compounds having four stereogenic centers with very high fidelity.

Full text of DOI:10.1021/ol300192p

ORGANIC
LETTERS
2012
Vol. 14, No. 5
1310–1313
Multicatalytic Asymmetric Synthesis of
Complex Tetrahydrocarbazoles via a
DielsꢀAlder/Benzoin Reaction Sequence
‡
‡
‡
,†,‡
ꢀ
Yankai Liu, Manuel Nappi, Eduardo C. Escudero-Adan, and Paolo Melchiorre*
ꢀ
´
ICREA - Institucio Catalana de Recerca i Estudis Avanc-ats Pg. Lluıs Companys, 23
08010 Barcelona, Spain, and ICIQ - Institute of Chemical Research of Catalonia Av.
Paısos Catalans 16, 43007 Tarragona, Spain
¨
pmelchiorre@iciq.es
Received January 24, 2012
ABSTRACT
Expanding upon the recently developed aminocatalytic asymmetric indole-2,3-quinodimethane strategy, a straightforward synthesis of
structurally and stereochemically complex tetrahydrocarbazoles has been devised. The chemistry’s complexity-generating power was further
harnessed by designing a multicatalytic, one-pot DielsꢀAlder/benzoin reaction sequence to stereoselectively access trans-fused tetracyclic
indole-based compounds having four stereogenic centers with very high fidelity.
Polycyclic indole architectures,1 including tetrahydro-
carbazole derivatives,1eꢀh are key structural elements in
many natural alkaloids and synthetic biologically active
compounds. There is thus great interest in developing
catalytic asymmetric methodologies to efficiently access
these privileged molecular scaffolds.1d,2
with multiple stereocenters using nontraditional dis-
connections.3 The chemistry used the intrinsic synthetic
power of the DielsꢀAlder reaction to provide predictable
and single-step access, from simple starting materials, to
stereochemically dense cyclohexenyl rings adorned with
different molecular architectures (Figure 1). The success of
this indole-2,3-quinodimethane strategy4 reliedonthe in situ
generation of intermediates II, reactive diene species that
were never before used in a catalytic approach. Upon con-
densation with enal 1 and the formation of the expected
electrophilic iminium ion intermediate I, the chiral sec-
ondary aminocatalyst A (diphenylprolinol silylether)5
induced a tautomerization event toward the reactive
species II, while selectively directing the [4 þ 2] cyclo-
addition reaction with suitable dienophiles toward a
highly enantioselective pathway.
Recently, our laboratory discovered a straightforward
synthetic route to complex tetrahydrocarbazole derivatives
‡ ICIQ.
† ICREA.
(1) (a) Indoles; Sundberg, R. J., Ed.; Academic Press: San Diego, CA,
1996. (b) Kleeman, A.; Engel, J.; Kutscher, B.; Reichert, D. Pharmaceutical
Substances, 4th ed.; Thieme: New York, 2001. (c) Cacchi, S.; Fabrizi, G.
Chem. Rev. 2005, 105, 2873. (d) Bandini, M; Eichholzer, A. Angew.
ꢀ
Chem., Int. Ed. 2009, 48, 9608. (e) Bonjoch, J.; Sole, D. Chem. Rev. 2000,
€
100, 3455. (f) Knolker, H. J.; Reddy, K. R. Chem. Rev. 2002, 102, 4303.
(g) Kato, D.; Sasaki, Y.; Boger, D. L. J. Am. Chem. Soc. 2010, 132, 3685.
(h) Jones, S. B.; Simmons, B.; Mastracchio, A.; MacMillan, D. W. C.
Nature 2011, 475, 183.
Recognition of the directness and versatility of the ami-
nocatalytic indole-2,3-quinodimethane strategy for the
stereocontrolled annulations of indole systems prompted
(2) Asymmetric catalytic preparation of tetrahydrocarbazoles: (a)
ꢀ
Tan, B.; Hernandez-Torres, G.; Barbas, C. F., III. J. Am. Chem. Soc.
2011, 133, 12354. (b) Zhu, X.-Y.; An, X.-L.; Li, C.-F.; Zhang, F.-G.;
Hua, Q.-L.; Chen, J.-R.; Xiao, W.-J. ChemCatChem 2011, 3, 679. (c)
Wang, X.-F.; Chen, J.-R.; Cao, Y.-J.; Cheng, H.-G.; Xiao, W.-J. Org.
Lett. 2010, 12, 1140. (d) Gioia, C.; Hauville, A.; Bernardi, L.; Fini, F.;
Ricci, A. Angew. Chem., Int. Ed. 2008, 47, 9236.
(3) Liu, Y.-K.; Nappi, M.; Arceo, E.; Vera, S.; Melchiorre, P. J. Am.
Chem. Soc. 2011, 133, 15212.
(4) Magnus, P.; Gallagher, T.; Brown, P.; Pappalardo, P. Acc. Chem.
Res. 1984, 17, 35.
(5) (a) Marigo, M.; Wabnitz, T. C.; Fielenbach, D.; Jørgensen, K. A.
Angew. Chem., Int. Ed. 2005, 44, 794. (b) Hayashi, Y.; Gotoh, H.;
Hayashi, T. Angew. Chem., Int. Ed. 2005, 44, 4212.
r
10.1021/ol300192p
Published on Web 02/16/2012
2012 American Chemical Society

DielsꢀAlder/benzoin reaction sequence. The chemistry pro-
vides straightforward access to highly functionalized tetra-
hydrocarbazoles 3 and 4bearing multiple stereogenic centers
with extremely high regio-, diastereo-, and enantioselectivity.
Scheme 1. Design Plan for the Multicatalytic Asymmetric
DielsꢀAlder/Cross-Benzoin Reaction Sequence to trans-Fused
Tetracyclic Products (EWG: Electron Withdrawing Group)
Figure 1. Aminocatalytic indole-2,3-quinodimethane strategy
(previous work used nitroolefins and methyleneindolinones as
the dienophiles, see ref 3).
us to further investigate its synthetic potential. In particular,
we wondered if the complexity-generating power of the
chemistry could be further expanded by integration into a
multicatalytic reaction sequence. Tandem or cascade reac-
tions have recently been recognized as powerful tools for
delivering dramatic increases in molecular and stereochem-
ical complexity via single-pot operations, without the need
for intermediate workup or purification.6 In addition, the
seminal studies by Rovis,7 Jørgensen,8 and Enders9 have
shown that it is possible to combine the catalytic activity of
secondary amines of type A and N-heterocyclic carbenes
(NHCs, catalysts of type B in Scheme 1) to stereoselectively
access complex molecules.6a Given the compatibility of
these two catalysts, we envisioned a direct way to stereo-
selectively access trans-fused tetracyclic indole-based com-
pounds 4 by combining the aminocatalytic-driven in situ
formation of indole-2,3-quinodimethane intermediates II
with an NHC-promoted intramolecular benzoin condensa-
tion (Scheme 1).10 For this to be feasible, we needed to use a
well-designed dienophile 2 in the first aminocatalytic step.
The multicatalytic system would only be operative if we
included a substrate bearing a keto moiety, the chemical
handle that is essential for benzoin condensation.
The versatility of the aminocatalytic indole-2,3-quino-
dimethane strategy was initially tested against the enal 1a/
trans-1,2-dibenzoylethylene 2a11 combination. We chose the
N-Boc protected 3-(2-methyl-indol-3-yl)acrylaldehyde 1a
because of its previously established ability to coax the
in situ formation of the key indole-2,3-quinodimethane
intermediate II upon condensation with the aminocatalyst
A (20 mol %). The commercially available 2a was selected
as the dienophile. This is because it bears the chemical
handle needed to develop the envisioned multicatalytic
strategy. Table 1 summarizes selected optimization studies
on the DielsꢀAlder process, which led to the tetrahydro-
carbazole 3a. Extensive details are reported in the Support-
ing Information (Tables S1ꢀS3).
Initial results confirmed that the indole-2,3-quinodi-
methane strategy, driven by catalyst A, can be successfully
extended to other dienophile classes, while maintaining its
inherent ability to rapidly build up complex frameworks
from simple starting materials and with high stereoselec-
tivity. Evaluation of the standard reaction parameters
revealed that both the reaction medium and the acidic
additive were crucial factors for improving the catalysis.
The best results were achieved when using toluene and a
catalytic system made by mixing equimolar amounts of
amine A with the bulky 2,4,6-trimethylbenzoic acid
(TMBA, entry 5). Importantly, the reaction temperature
increased the catalyst turnover while minimally affecting
the stereoselectivity of the cycloaddition process (entry 6).
From a synthetic standpoint, it is worth mentioning that
the use of the antipode of catalyst A showed the same
catalytic profile and stereochemical outcome, granting
access to each product enantiomer individually (entry 7).
Here we describe the successful realization of this idea,
based on the extension of the indole-2,3-quinodimethane
strategy to include novel classes of keto-containing dieno-
philes 2 and the optimization of an unprecedented one-pot
(6) For reviews: (a) Grossmann, A.; Enders, D. Angew. Chem., Int.
Ed. 2012, 51, 314. (b) Ambrosini, L. M.; Lambert, T. H. ChemCatChem
2010, 2, 1373. (c) Grondal, C.; Jeanty, M.; Enders, D. Nat. Chem. 2010,
2, 167. (d) Zhou, J. Chem.;Asian J. 2010, 5, 422. Selected examples: (e)
Jiang, H.; Elsner, E.; Jensen, K. L.; Falcicchio, A.; Marcos, V.; Jørgen-
sen, K. A. Angew. Chem., Int. Ed. 2009, 48, 6844. (f) Simmons, B.; Walji,
A. M.; MacMillan, D. W. C. Angew. Chem., Int. Ed. 2009, 48, 4349.
(7) (a) Lathrop, S. P.; Rovis, T. J. Am. Chem. Soc. 2009, 131, 13628.
(b) Ozboya, K. E.; Rovis, T. Chem. Sci. 2011, 2, 1835.
(8) Jacobsen, C. B.; Jensen, K. L.; Udmark, J.; Jørgensen, K. A. Org.
Lett. 2011, 13, 4790.
(9) (a) Enders, D.; Grossmann, A.; Huang, H.; Raabe, G. Eur. J. Org.
Chem. 2011, 4298. See also: (b) Hong, B.-C.; Dange, N. S.; Hsu, C.-S.;
Liao, J.-H.; Lee, G.-H. Org. Lett. 2011, 13, 1338.
(10) Previous studies showed how carbene catalysis could be com-
bined with distinct aminocatalytic activation modes driven by amine A.
References 7a, 8, and 9a deal with iminium ion activation. Reference 7b
deals with enamine activation. Here, we provide the first demonstration
that the reactivity of an extended enamine (the trienamine intermediate
II) can also be integrated into a multicatalytic process.
(11) Wang, Y.; Li, H.-M; Wang, Y.-Q.; Liu, Y.; Foxman, B. M.;
Deng, Li. J. Am. Chem. Soc. 2007, 129, 6364.
Org. Lett., Vol. 14, No. 5, 2012
1311

results. There appears to be significant tolerance toward
structuralandelectronicvariationsof substitutionpatterns
on the dienophiles. Different substituents at the aromatic
moiety of the symmetric trans-1,2-dibenzoylethylene deriv-
atives were well-tolerated, regardless of their electronic
properties. The corresponding tetrahydrocarbazoles 3aꢀd
were obtained in good yield and very high levels of stereo-
control (up to 8:1 dr, 97ꢀ99% ee). We also explored the
influenceofanunsymmetrical substitution pattern onboth
sides of the dienophile double bond. We successfully
extended the chemistry to include nonsymmetrical trans-
benzoylacrylic esters (R = aryl, EWG = ester moiety),
achieving analmostperfect level of chemo-, diastereo-, and
enantioselectivity (products 3eꢀl in Figure 2).
Table 1. Selected Optimization Studiesa
yieldb
(%)
eed
entry
solvent
acid
drc
nd
(%)
1
toluene
toluene
DCE
ꢀ
<5
38
48
42
52
84
85
nd
97
94
97
96
98
98f
2
PhCO2H
PhCO2H
o-NO2-PhCO2H
TMBA
5.6:1
6.3:1
8:1
3
The catalytic system also maintained its efficiency when
applied on a synthetically useful scale (1.0 mmol), leading
to the formation of the adducts 3c and 3e with comparable
yield and optical purity (results between brackets refer to a
0.1 mmol scale). We also found that the stereochemistry of
the cycloadducts 3 appeared to be insensitive to the double
bond geometry of 2. Indeed, starting from a trans- or cis-
configured dienophile, the corresponding tetrahydrocar-
bozole 3f was forged with the same relative and absolute
configuration. This is because, under the reaction condi-
tions, a scrambling of the double bond occurs prior to
the cycloaddition.12 It is noteworthy that the indole-2,
3-quinodimethane strategy can also be applied to benzoyl
acrylonitrile (EWG = CN), with the corresponding pro-
duct 3m formed with perfect regioselectivity. Finally, we
4
toluene
toluene
toluene
toluene
5
8:1
6e
7f
TMBA
8:1
TMBA
8:1
a Reactions performed mixing 1a (0.12 mmol), 2a (0.1 mmol), (R)-A
(0.02 mmol), and acid (0.02 mmol) in 0.2 mL of the solvent at ambient
temperature for 16 h. b Yield of the isolated compound 3a. c Determined by
1H NMR analysis of the crude reaction mixture. d Determined by HPLC on
a chiral stationary phase. e Reaction performed at 40 °C over 48 h. f The (S)
enantiomer of amine Awas used, leading to the opposite antipode of product
3a. DCE = 1,2-dichloroethane, TMBA = 2,4,6-trimethylbenzoic acid.
The conditions in entry 6 were selected to examine the
scope of the DielsꢀAlder process by evaluating a variety of
keto-containing dienophiles 2. Figure 2 summarizes the
a
Figure 2. Extending the aminocatalytic indole-2,3-quinodimethane strategy to new classes of dienophiles. Unless noted otherwise,
reactions performed at 40 °C with 1 (0.12 mmol), 2 (0.1 mmol), A, and TMBA (0.02 mmol) in toluene (0.2 mL) for 48 h. Yields refer to
the sum of the isolated diastereoisomers 3 and reflect the degree of conversion. bYield of the isolated single, major diastereoisomer.
c1 mmol scale reaction: results between brackets refer to 0.1 mmol scale reaction. dReaction at 50 °C. e2,6-Bis(trifluoromethyl) benzoic
f
acid (20 mol %) was used. Yield and ee were determined after in situ homologation of aldehydes 3hꢀi,lꢀm with a Wittig reagent
(PPh3dCHCO2Et, 0.14 mmol, rt, 16 h). gReaction at 60 °C.
1312
Org. Lett., Vol. 14, No. 5, 2012

showed that different substituents on the indole core of
enal 1 were well tolerated. Substitution at the 5-position
efficiently provided diene precursors for the highly stereo-
selective DielsꢀAlder reaction with trans-benzoylacrylic
esters, leading to products 3k and 3l.
Scheme 2. Multicatalytic DielsꢀAlder/Cross-Benzoin Reaction
Sequence to trans-Fused Tetracyclic Products
Figure 3. Single crystal X-ray diffraction data for 4b.
densely functionalized trans-fused tetracyclic indole-based
compounds 4 with four stereogenic centers with very high
fidelity (Scheme 2). While symmetric dibenzoylethylene
derivatives worked well in the multicatalytic sequence,
trans-benzoylacrylic esters led to complex product mix-
tures. This is probably because of a transesterification
event between the ester group and the newly generated
OH group under basic conditions.
An X-ray crystal structure analysis of the dichloro-
substituted cycloadduct 4b unambiguously allowed the deter-
mination of the stereochemical outcome of the multicatalytic
process (Figure 3).15 An exo-selective DielsꢀAlder process3,14b
is followed by a completely stereoselective benzoin condensa-
tion, proceeding through a substrate-directed pathway.16
The amino-catalytic indole-2,3-quinodimethane strat-
egy can rapidly build up complex frameworks from simple
starting materials. We have shown how its potential can be
expanded to include a variety of different dienophiles. The
implementation of a multicatalytic, one-pot DielsꢀAlder/
benzoin reaction sequence, based upon the unprecedented
combination of trienamine and carbene catalysis, led to the
highly stereoselective preparation of complex tetrahydro-
carbazole derivatives.
With suitable conditions elaborated for a productive
pericyclic step, we turned our attention to implementing a
multicatalytic DielsꢀAlder/benzoin reaction sequence.
We initially tried to have both the aminocatalyst A and
the NHC precursor B from the outset of the process.
However, the DielsꢀAlder reaction was very sluggish in
the presence of a base, which was required to in situ form
the active carbene catalyst.13 Consequently, the achiral
triazolium salt B and sodium acetate were added in a
sequential one-pot manner at an elevated temperature
after completion of the DielsꢀAlder reaction. The result-
ing multicatalytic process involved an amino-catalyzed
pericylic pathway under trienamine activation14 of enals
1, followed by a N-heterocyclic carbene-catalyzed intra-
molecular crossed-benzoin reaction. It led directly to the
Acknowledgment. This work was supported by the In-
stitute of Chemical Research of Catalonia (ICIQ) Founda-
tion and by MICINN (Grant CTQ2010-15513). Y.L. is
grateful to the European Union for a Marie Curie fellow-
ship (Grant FP7-PEOPLE-2010-IIF n: 273088). The
authors thank Dr. Jamin Krinsky (Universitat Rovira i
Virgili, Tarragona) for theoretical models.
(12) NMR spectroscopy confirmed a quick isomerization event, with
the trans dienophile being the final active species of the pericyclic
process.
(13) The absence of an acid cocatalyst resulted in a very sluggish
DielsꢀAlder reaction; see Table 1, entry 1. The complete progress
towards the optimized conditions for the multicatalytic sequence is
detailed in Table S4 within the Supporting Information.
(14) DielsꢀAlder reactions via trienamine intermediates: (a) Jia,
Z.-J.; Jiang, H.; Li, J.-L.; Gschwend, B.; Li, Q.-Z.; Yin, X.; Grouleff, J.;
Chen, Y.-C.; Jørgensen, K. A. J. Am. Chem. Soc. 2011, 133, 5053. (b) Jia,
Z.-J.; Zhou, Q.; Zhou, Q.-Q.; Chen, P.-Q.; Chen, Y.-C. Angew. Chem.,
Int. Ed. 2011, 37, 8638. (c) Jiang, H.; Gschwend, B.; Albrecht, L.;
Hansen, S. G.; Jørgensen, K. A. Chem.;Eur. J. 2011, 33, 9032.
(15) Crystallographic data for compound 4b are available free of
charge from the Cambridge Crystallographic Data Centre, accession
number CCDC 862664.
Supporting Information Available. Complete experi-
mental procedures, compound characterization, HPLC
traces, and NMR spectra (PDF). This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
(16) The optical purity of products 4 reflects the stereoselectivity of
the DielsꢀAlder process, leading to precursors 3.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 5, 2012
1313