Remarkably reactive dihydroindoloindoles via palladium-catalysed dearomatisation

Article abstract of DOI:10.1039/c0cc05033d

Palladium-catalysed dearomatisation reactions allow access to a previously unknown class of indoloindole heterocycle: 5,10b-dihydroindolo[2,3-b]indoles. The highly reactive nature of these compounds is demonstrated by their facile reactions with water and with hydride, alkyl, aryl and allyl-based organometallic nucleophiles.

Full text of DOI:10.1039/c0cc05033d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 3649–3651
www.rsc.org/chemcomm
COMMUNICATION
Remarkably reactive dihydroindoloindoles via palladium-catalysed
dearomatisationw
Robin B. Bedford,* Natalie Fey, Mairi F. Haddow and Rosalind F. Sankey
Received 18th November 2010, Accepted 28th January 2011
DOI: 10.1039/c0cc05033d
Palladium-catalysed dearomatisation reactions allow access to a
previously unknown class of indoloindole heterocycle: 5,10b-
dihydroindolo[2,3-b]indoles. The highly reactive nature of these
compounds is demonstrated by their facile reactions with water
and with hydride, alkyl, aryl and allyl-based organometallic
nucleophiles.
Catalytic dearomatisation reactions (Scheme 1) offer the
expediency of transition-metal catalysed biaryl cross-coupling
processes while increasing the extent of topological complexity
via the introduction of chiral, quaternary sp3 carbons
(Scheme 1).
There are currently very few examples of transition metal-
catalysed dearomatisation processes being exploited in synthesis
and those examples are limited to the formation of stable
products.1,2 By contrast our interest lies in the synthesis of
novel classes of molecules formed by catalytic dearomatisation
that are ordinarily far too reactive to isolate. These highly
labile species may reveal and facilitate unusual reactivities that
Scheme 2 Conditions: 2a (0.5 mmol), Pd(OAc)2 (5 mol%), SiPr
(14 mol%), NaOtBu (1.5 mmol), toluene (5 ml), 100 1C, 16 h.
can in turn be exploited in novel synthetic pathways. We
recently demonstrated that this approach can be used successfully
in the isolation of very reactive 4a-alkyl-4aH-carbazoles,3
synthesis of a novel and highly reactive class of indoloindole,
5,10b-dihydroindolo[2,3-b]indoles, 1.5
which had previously only been postulated as reactive inter-
Scheme 2 illustrates the conversion of the precursor
2a—synthesised by the Buchwald–Hartwig amination of
2-chloroaniline with the appropriate 2-bromoindole—to the
desired product 5,10b-dimethyl-5,10b-dihydroindolo[2,3-b]-
mediates formed during photolysis or flash vacuum pyrolysis
experiments or isolated as electronically-stabilised benzannulated
congeners.1b,4
We wished to show that palladium-catalysed dearomatisation
could be exploited to generate previously unknown examples
of reactive heterocycles and report here its application to the
1
indole, (ꢀ)1a, in 90% spectroscopic yield (as determined by H
NMR spectroscopy) using sodium tert-butoxide as base. The
catalyst used was formed in situ from palladium acetate and the
N-heterocyclic carbene ligand SIPr, as its pentafluorobenzene
adduct.6 The intermediate formed by oxidative addition of 2a to
the palladium can undergo dearomatisation by at least two
distinct routes: (a) an electrophilic or (b) a Heck pathway.
The Heck pathway seems less likely on the grounds of the
high ring-strain of the trans-fused intermediate 3 which would
be formed by the required syn-carbopalladation of the indole
double bond. Furthermore the reaction was found to be
strongly base dependent, as would be anticipated if the
dearomatisation was indeed triggered by deprotonation of the
NH function.1b,3 Of the range of bases explored,7 only NaOtBu
and Cs2CO3 gave the desired product, the latter in somewhat
diminished yield (60%), militating against a Heck-like process
where all the bases tested are known to be effective.
Scheme 1 Catalytic dearomatisation.
School of Chemistry, University of Bristol, Cantock’s Close, Bristol,
BS8 1TS, UK. E-mail: r.bedford@bristol.ac.uk
w Electronic supplementary information (ESI) available: Experimental,
X-ray crystal data, spectroscopic data and details of the computational
method. CCDC 789828–789829. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0cc05033d
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3649–3651 3649

View Article Online
Scheme 3 Conditions: 2 (0.5 mmol), Pd(OAc)2 (5 mol%), SiPr
(14 mol%), NaOtBu (1.5 mmol), toluene (5 ml), 100 1C, 16 h.
Spectroscopic yields, determined by 1H NMR (1,3,5-MeOC6H4,
internal standard).
Scheme 4 Reactivity of 5,10b-dihydroindolo[2,3-b]indoles.
We next examined the extension of the catalytic dearomatisation
methodology to other representative products 1 (Scheme 3).
As can be seen both alkyl and aryl substitution was tolerated
at the site of quaternisation, as were both electron-donating
and withdrawing groups on the aryl chloride fragment.
All the products 1 proved to be highly moisture sensitive,
undergoing facile hydrolysis with adventitious water, and in all
cases they were produced as oils containing varying
proportions of 4. NMR and high-resolution mass spectroscopic
data fully supported the structures of the indoloindoles 1,8,9
however the best confirmation of the identity of the labile
species 1 came from their subsequent reactions which centered
around their propensity to undergo nucleophilic attack at the
iminic carbon.
Fig. 1 X-Ray crystal structure of 4a, thermal ellipsoids set at 30%
probability level.
Thus the indoloindoles 1 could be cleanly converted to
3,30-disubstituted oxindoles 4, compounds with a range of
interesting biological properties,10 by simply subjecting them
to column chromatography on silica (Scheme 4). In addition
1a underwent reactions both with superhydride and a range of
organometallic-based alkyl, aryl and allyl nucleophiles to
generate the tetrahydroindolo[2,3-b]indoles 5. To the best of
our knowledge these are the first examples of tetrahydroindolo-
[2,3-b]indoles with alkyl and dialkyl substitution across the
ring junction, indeed there has only been one other reported
synthesis of tetrahydroindolo[2,3-b]indoles.11 In all cases the
reactions proceeded with strict cis diastereoselectivity across
the C–C ring junction.
Fig. 2 X-Ray crystal structure of 5b, thermal ellipsoids set at 30%
probability level.
greater ring strain associated with the introduction of
benzannulation, which would be relieved on removal of the
imine bond. Indeed DFT calculations show that the model of
4a (4a*) lies 31.8 kcal molꢁ1 lower in energy than the reactants,
while for the equivalent hydrolysis of tetrahydropyrroloindole
The X-ray crystal structures of representative hydrolysis
and alkylation products, 4a and 5b, were determined,
demonstrating unequivocally the site of C–C bond formation
during the catalytic dearomatisation process (Fig. 1 and 2).
The facile reactivity of 1 with water is in stark contrast with
the reactivity of known tetrahydropyrrolo[2,3-b]indole analogues,
as exemplified by flustramine C,12 which is moisture stable.
This difference in reactivity is presumably large due to the far
6* this energy difference is reduced to 25.6 kcal molꢁ1 8,13
.
It may be anticipated that the greater ring strain would lead
to a longer and perhaps weaker CQN bond and thus more
facile hydrolysis, however calculations showed the C4–N5
c
3650 Chem. Commun., 2011, 47, 3649–3651
This journal is The Royal Society of Chemistry 2011

View Article Online
bond lengths to be very similar in the models 1a* and 6*
(1.291 and 1.297 A respectively).8 In fact it appeared that
changes in geometry around N3 on hydrolysis play a far more
important role.14
2 For an example of an organocatalytic dearomatisation see:
N. T. Vo, R. D. M. Pace, F. O’Hara and M. J. Gaunt, J. Am.
Chem. Soc., 2008, 130, 404.
3 R. B. Bedford, C. P. Butts, M. F. Haddow, R. Osborne and
R. F. Sankey, Chem. Commun., 2009, 4832.
4 (a) J. J. Kulagowski, C. J. Moody and C. W. Rees, J. Chem. Soc.,
Chem. Commun., 1982, 548; (b) J. J. Kulagowski, C. J. Moody and
C. W. Rees, J. Chem. Soc., Perkin Trans. 1, 1985, 2725;
(c) J. J. Kulagowski, G. Mitchell, C. J. Moody and C. W. Rees,
J. Chem. Soc., Chem. Commun., 1985, 650; (d) J. J. Kulagowski,
C. J. Moody and C. W. Rees, J. Chem. Soc., Perkin Trans. 1, 1985,
2733.
5 For related fused pyrroles see: L. Benati, G. Bencivenni,
R. Leardini, M. Minozzi, D. Nanni, R. Scialpi, P. Spagnolo,
G. Zanardi and C. Rizzoli, Org. Lett., 2004, 6, 417.
In summary, we have demonstrated that palladium-
catalysed dearomatisation can be exploited in the synthesis
of a previously unknown and highly reactive class of indolo-
indole which in turn serves as a starting point for the synthesis
of heterocycles with non-planar, densely functionalised cores. We
are currently exploring the extension of this novel catalytic
dearomatisation to the synthesis of a range of heterocyles.
6 G. W. Nyce, S. Csihony, R. M. Waymouth and J. L. Hedrick,
Chem.–Eur. J., 2004, 10, 4073.
7 NaOAc, K3PO4 and Cy2NMe gave no dearomatised product only
starting materials and some unidentified decomposition products.
8 See ESIw.
9 The 13C NMR spectra showed peaks in the range 189.6–192.3 ppm
corresponding to the iminic carbons, while the fused quaternary
benzylic carbons were seen in the range 59.7–60.6 ppm.
10 (a) B. M. Trost, N. Cramer and H. Bernsmann, J. Am. Chem. Soc.,
2007, 129, 3086; (b) H. Lin and S. J. Danishefsky, Angew. Chem.,
Int. Ed., 2003, 42, 36; (c) X. Z. Wearing and J. M. Cook, Org. Lett.,
2002, 4, 4237; (d) N. H. Greig, X. F. Pei, T. T. Soncrant,
D. K. Ingram and A. Brossi, Med. Res. Rev., 1995, 15, 3.
11 P. Magnus and R. Turnbull, Org. Lett., 2006, 8, 3497.
12 T. Lindel, L. Bra
283.
13 Transition state searches lay outside the scope of this work, as we
have not yet established the detailed mechanism.
¨ ¨
uchle, G. Golz and P. Bohrer, Org. Lett., 2007, 9,
We thank Dr Robert Osborne (AZ) for stimulating
discussions and EPSRC and AZ for funding (industrial CASE
studentship for RFS). NF thanks the EPSRC for the award of
an Advanced Research Fellowship (EP/E059376/1).
14 While the C2–N3–C4 angle increased by almost 51 on formation of
4a* from 1a*, the equivalent structural change was somewhat
smaller (3.41) on going from 6* to 7*. Furthermore, the N3 of
4a* is essentially trigonal planar, this nitrogen atom is more
pyramidal in the starting material 1a* (S(angles) = 352.81). In
6* and 7* these sums were much closer (355.6 and 356.21,
respectively), suggesting that the geometry around this nitrogen
is less constrained and that relief of strain on breaking the CQN
bond in 6* would play much less of a role than in the equivalent
hydrolysis of 1a*.
Notes and references
1 For examples see: (a) K. Funabshi, H. Ratni, M. Kanai and
M. Shibasaki, J. Am. Chem. Soc., 2001, 123, 10784;
(b) J. Garcıa-Fortanet, F. Kessler and S. L. Buchwald, J. Am.
´
Chem. Soc., 2009, 131, 6676; (c) B. Peng, X. Feng, X. Zhang,
S. Zhang and M. Bao, J. Org. Chem., 2010, 75, 2619.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3649–3651 3651