Copper-mediated construction of spirocyclic bis-oxindoles via a double C-H, Ar-H coupling process

Article abstract of DOI:10.1021/ol5024129

A double C-H, Ar-H coupling process for the conversion of bis-anilides into spirocyclic bis-oxindoles, enabling the concomitant formation of two all-carbon quaternary centers at oxindole 3-positions in a diastereoselective manner, is described. The optimum cyclization conditions utilize stoichiometric Cu(OAc)₂·H₂O/KOtBu in DMF at 110 °C and have been applied to prepare a range of structurally diverse bis-spirooxindoles in fair to good yields (28-77%); the method has also been extended to prepare bis-oxindoles linked by a functionalized acyclic carbon chain.

Full text of DOI:10.1021/ol5024129

Letter
pubs.acs.org/OrgLett
Copper-Mediated Construction of Spirocyclic Bis-oxindoles via a
Double C−H, Ar−H Coupling Process
Pauline Drouhin, Timothy E. Hurst, Adrian C. Whitwood, and Richard J. K. Taylor*
Department of Chemistry, University of York, Heslington, York YO10 5DD, U.K.
S
* Supporting Information
ABSTRACT: A double C−H, Ar−H coupling process for the conversion of bis-anilides into spirocyclic bis-oxindoles, enabling
the concomitant formation of two all-carbon quaternary centers at oxindole 3-positions in a diastereoselective manner, is
described. The optimum cyclization conditions utilize stoichiometric Cu(OAc)2·H2O/KOtBu in DMF at 110 °C and have been
applied to prepare a range of structurally diverse bis-spirooxindoles in fair to good yields (28−77%); the method has also been
extended to prepare bis-oxindoles linked by a functionalized acyclic carbon chain.
ver the past decade, there has been a significant
resurgence of interest in oxindoles, as these structures
represent validated targets in the search for new drug
candidates and form the cornerstone of numerous alkaloids
of biological interest.1 More recently, bis-oxindoles have
attracted considerable attention (Figure 1). For example,
ing multiple spiro-quaternary carbon centers. The majority of
approaches rely on the linking together of preformed
oxindoles,2−4 a strategy most beautifully illustrated by Barbas
who developed an organocatalytic asymmetric Michael
addition/aldol cascade reaction between 3-substituted oxin-
doles and methyleneindolinones which proceeds with excellent
diastereo- and enantiocontrol (Scheme 1, eq 1).6 Related
variants have subsequently been reported7 but all commence
with preformed oxindoles and all produce bis-oxindole
O
Scheme 1. Strategies for the Synthesis of Bis-oxindoles
Figure 1. Examples of bis-oxindole targets.
Natura (1) is representative of a family of isoindigo-based
anticancer agents (CDK inhibitors)2 and compound 2 is typical
of a range of dispirooxindole-pyrrolidine derivatives recently
shown to possess significant antibacterial and anticancer
activities (against A549 human lung adenocarcinoma).3 More-
over, bis-oxindoles have long been employed as precursors of
bis(pyrroloindoline) alkaloids,4 most recently with compound 3
being used as a cornerstone for the synthesis of a diverse range
of cyclotryptamine alkaloids.4d
Many synthetic strategies have been established to access the
oxindole motif,5 but only limited examples have been reported
to date on bis-oxindoles, probably because of their highly
functionalized polycyclic skeletons, particularly those contain-
Received: August 14, 2014
Published: September 8, 2014
© 2014 American Chemical Society
4900
dx.doi.org/10.1021/ol5024129 | Org. Lett. 2014, 16, 4900−4903

Organic Letters
Letter
products linked at the 3,3-positions by another 5-membered
ring generated in a formal [3 + 2]-cyclization process.8
Our objective was to establish a more general route to bis-
oxindoles (Scheme 1, eq 2) which would utilize readily
accessible bis-anilide precursors9 and would be applicable to the
formation of a range of diverse bis-oxindole products with, for
the first time, great variability in the linking central core units.
As shown, the plan was to utilize a copper(II)-mediated bis-
anilide cyclization approach (a formal C−H, Ar−H coupling)
based on the chemistry devised for the preparation of
oxindoles10,11 and related heterocycles12 by the groups of
Taylor and Kundig in 2009. Herein, we wish to disclose the
̈
Figure 2. Crystal structure of 5a (50% probability ellipsoids).
success of this approach to access a range of bis-spirooxindoles
featuring central core units of different ring sizes and, in
addition, functionalized acyclic linker units.
substrate scope using a range of substituted bis-anilides 4.15
First we ensured that the procedure was compatible with N-
benzyl protection and found that adduct 5b was formed in 57%
yield. Substitution of the aromatic rings was studied next, and
both 4-methyl- and 4-methoxy-substitution was well tolerated
giving 5c and 5d, respectively. Unsymmetrical bis-oxindoles
were also prepared with either differential ring substitution (5e)
or differential N-protection (5f). Variation of the central ring
size was also explored.15 Thus, a cyclohexanone (5g) and a
substituted cyclohexanone example (5h) were prepared, as
were a 7-membered ring-containing bis-oxindole (5i, obtained
in 77% yield) and a benzo-fused cycloheptanone example (5j).
The yields of the cyclization products varied (28−77%), but it
should be noted that all of the procedures in Scheme 2 used the
standard conditions developed in Table 1 and none were
optimized. It should also be noted that all of products in
Scheme 2 were obtained as single trans-diastereoisomers.
The cyclopentanone 2,5-dicarboxamide 4a was chosen as the
bis-anilide for preliminary studies (Table 1). Compound 4a was
Table 1. Optimization of the Reaction Conditions
solvent
(temp)
time
(h)
entry
base
Cu source
yield
<5%
1
−
−
−
Cu(OAc)2·H2O
(1.0 equiv)
mesitylene
(170 °C)
toluene
(100 °C)
toluene
(80 °C)
DMF
(110 °C)
0.5
0.5
3
2
3
4
Cu(OAc)2·H2O
(1.0 equiv)
24%
17%
67%
Cu(OAc)2·H2O
(1.0 equiv)
Scheme 2. Bis-spirooxindole Substrate Scope
a
KOtBu
(2.2 equiv)
Cu(OAc)2·H2O
(2.0 equiv)
0.25
a
In a control experiment carried out using KOtBu (2.2 equiv) but
1
without Cu(OAc)2·H2O, no product was observed in the H NMR
spectrum of the crude reaction mixture. Instead, residual starting
material and products from amide hydrolysis as well as decomposition
were observed.
prepared in two steps by amide coupling between adipic acid
and N-methylaniline followed by a novel ring closure using
carbonyldiimidazole (CDI; see Supporting Information for
details). The thermodynamically more stable trans-bis-anilide
diastereoisomer was the only product of the reaction
(confirmed by X-ray crystallography).13 The double spirocyc-
lization was then investigated, initially using Cu(OAc)2·H2O in
mesitylene at 170 °C, conditions optimized for the formation of
simple oxindoles.10b,c However, only traces of cyclized product
5a were detected under these conditions (Table 1, entry 1).
Carrying out the reaction in toluene at reflux gave a useful 24%
yield (entry 2) but further reductions in temperature resulted in
lower yields, even after extended reaction times (e.g., entry 3).
However, changing to Cu(OAc)2·H2O and KOtBu in DMF
(conditions employed in our original study10a) gave the desired
spirocyclic bis-oxindole 5a in a gratifying 67% yield (entry 4).
This double cyclization proceeded with complete diasterose-
lectivity to give only the trans-diastereoisomer illustrated, in
which the two oxindole units are orthogonal (confirmed by X-
ray crystallographic analysis, Figure 2).14
Having established successful conditions for the double
cyclization on the model system 4a, we went on to test the
4901
dx.doi.org/10.1021/ol5024129 | Org. Lett. 2014, 16, 4900−4903

Organic Letters
Letter
To further demonstrate the scope and versatility of the
double anilide cyclization procedure, attention switched to the
use of acyclic linker units. Initial studies were carried out on bis-
anilide 9, readily prepared as shown in Scheme 3. Thus, acid
Scheme 4. Bis-oxindole Synthesis with Di-ketone Linkers
Scheme 3. Bis-oxindole Synthesis with an Acyclic Linker
chloride 7 was synthesized from the respective ester anilide 6 in
89% yield over two steps. The enolate of 8 was trapped with
acid chloride 7 and afforded bis-anilide 9 in 69% yield. In this
case, the strongly basic procedure was not required, and
treatment of precursor 9 with Cu(OAc)2·H2O (1 equiv) in
mesitylene at 170 °C for 30 min afforded the desired
bisoxindole 10 in 67% yield.16 The keto-linked bis-oxindole
10 was obtained as a mixture of diastereoisomers (1.5:1 trans/
cis) which were separable by column chromatography. The
relative configurations of the diastereoisomeric products were
determined by X-ray crystallography (Figure 3) indicating that
the trans-isomer 10 was the major product.17
Figure 4. Crystal structures of DL-12a (left) and meso-12a (right).
In conclusion, we have developed a concise strategy to access
a diverse range of spirocyclic bis-oxindoles using a one-pot,
double Cu(II)-mediated bis-anilide cyclization by double C−H,
Ar−H coupling. This method allows the installation of two all-
carbon quaternary centers at the oxindole 3-position in a
diastereoselective manner and great variability in the linking
central core units for the first time. The method has been
extended to prepare a number of bis-oxindoles linked by a
functionalized acyclic carbon chain. We are currently
investigating the use of chiral auxiliaries in these processes as
well as exploring applications in target synthesis.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and full spectroscopic data for all new
compounds are provided. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
Figure 3. Crystal structures of trans-10 (left) and cis-10 (right).
*E-mail: richard.taylor@york.ac.uk.
Notes
Having established that this Cu(II) method could be
employed to prepare bis-oxindoles with a keto-functionalized
one-carbon linker, we sought to further extend the reaction
scope (Scheme 4).
Compounds 11a−c were easily obtained from the corre-
sponding diacid chlorides using similar procedures to those
employed in Scheme 3. Cyclization again occurred efficiently
using Cu(OAc)2·H2O (1 equiv) in mesitylene at 170 °C, and
again mixtures of diastereoisomers were produced. The
diastereoisomers of bis-oxindole 12a, linked by a three-carbon
dicarbonyl chain, were separable, and the structures confirmed
by X-ray analysis (Figure 4).18 We also varied the linker chain
length introducing an aromatic ring (12b) and prepared the
adamantane-linked example 12c using a similar procedure.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the University of York Wild Fund for postgraduate
support (P.D.) and the EPSRC for postdoctoral funding
(T.E.H.; EP/J000124/1).
REFERENCES
■
(1) For reviews, see: (a) Cerchiaro, G.; da Costa Ferreira, A. M. J.
Braz. Chem. Soc. 2006, 17, 1473. (b) Peddibhotla, S. Curr. Bioact.
Compd. 2009, 5, 20.
(2) Arun, Y.; Bhaskar, G.; Balachandran, C.; Ignacimuthu, S.;
Perumal, P. T. Bioorg. Med. Chem. Lett. 2013, 23, 1839.
4902
dx.doi.org/10.1021/ol5024129 | Org. Lett. 2014, 16, 4900−4903

Organic Letters
Letter
(3) Wee, X. K.; Yeo, W. K.; Zhang, B.; Tan, V. B. C.; Lim, K. M.; Tay,
T. E.; Go, M.-L. Bioorg. Med. Chem. 2009, 17, 7562.
1016758), which can be obtained free of charge via www.ccdc.cam.
ac.uk/data_request/cif.
(4) (a) Link, J. T.; Overman, L. E. J. Am. Chem. Soc. 1996, 118, 8166.
(b) Movassaghi, M.; Schmidt, M. A. Angew. Chem., Int. Ed. 2007, 46,
3725. (c) Kim, J.; Ashenhurst, J. A.; Movassaghi, M. Science 2009, 324,
238. (d) Trost, B. M.; Osipov, M. Angew. Chem., Int. Ed. 2013, 52,
9176.
(18) X-ray data for meso-12a and dl-12a have been deposited with the
Cambridge Crystallographic Data Centre (meso-12a: CCDC 1013389,
DL-12a: CCDC 1013390), which can be obtained free of charge via
www.ccdc.cam.ac.uk/data_request/cif. See Supporting Information for
crystallographic details.
(5) For reviews, see: (a) Marti, C.; Carreira, E. M. Eur. J. Org. Chem.
2003, 2209. (b) Galliford, C. V.; Scheidt, K. A. Angew. Chem., Int. Ed.
2007, 46, 8748. (c) Trost, B. M.; Brennan, M. K. Synthesis 2009, 3003.
(d) Millemaggi, A.; Taylor, R. J. K. Eur. J. Org. Chem. 2010, 4527.
(e) Klein, J. E. M. N.; Taylor, R. J. K. Eur. J. Org. Chem. 2011, 6821.
(f) Singh, G. S.; Desta, Z. Y. Chem. Rev. 2012, 112, 6104.
(6) Tan, B.; Candeias, N. R.; Barbas, C. F., III. Nat. Chem. 2011, 3,
473.
(7) (a) Han, Y.-Y.; Chen, W.-B.; Han, W.-Y.; Wu, Z.-J.; Zhang, X.-M.;
Yuan, W.-C. Org. Lett. 2012, 14, 490. (b) Sun, W.; Zhu, G.; Wu, C.;
Hong, L.; Wang, R. Chem.Eur. J. 2012, 18, 6737. (c) Cao, Y.-M.;
Shen, F.-F.; Zhang, F.-T.; Wang, R. Chem.Eur. J. 2013, 19, 1184.
(d) Wu, H.; Zhang, L.-L.; Tian, Z.-Q.; Huang, Y.-D.; Wang, Y.-M.
Chem.Eur. J. 2013, 19, 1747. (e) Han, W.-Y.; Li, S.-W.; Wu, Z.-J.;
Zhang, X.-M.; Yuan, W.-C. Chem.Eur. J. 2013, 19, 5551. (f) Tan, F.;
Cheng, H.-G.; Feng, B.; Zou, Y.-Q.; Duan, S.-W.; Chen, J.-R.; Xiao,
W.-J. Eur. J. Org. Chem. 2013, 2071. (g) Wu, S.; Zhu, X.; He, W.;
Wang, R.; Xie, X.; Qin, D.; Jing, L.; Chen, Z. Tetrahedron 2013, 69,
11084. (h) Sun, J.; Xie, Y.-J.; Yan, C.-G. J. Org. Chem. 2013, 78, 8354.
(i) Sun, W.; Hong, L.; Zhu, G.; Wang, Z.; Wei, X.; Ni, J.; Wang, R.
Org. Lett. 2014, 16, 544.
(8) For an exception which produces cyclopropane-linked oxindoles,
see: (a) Noole, A.; Jarving, I.; Werner, F.; Lopp, M.; Malkov, A.;
̈
Kanger, T. Org. Lett. 2012, 14, 4922. (b) Noole, A.; Ose
̌
ka, M.; Pehk,
̈
T.; Oeren, M.; Jarving, I.; Elsegood, M. R. J.; Malkov, A. V.; Lopp, M.;
Kanger, T. Adv. Synth. Catal. 2013, 355, 829.
̈
(9) For earlier, less general, approaches starting from bis-anilides, see:
(a) Overman, L. E.; Paone, D. V.; Stearns, B. A. J. Am. Chem. Soc.
1999, 121, 7702. (b) Wang, J.; Yuan, Y.; Xiong, R.; Zhang-Negrerie,
D.; Du, Y.; Zhao, K. Org. Lett. 2012, 14, 2210. (c) Wu, H.; He, Y.-P.;
Xu, L.; Zhang, D.-Y.; Gong, L.-Z. Angew. Chem., Int. Ed. 2014, 53,
3466.
(10) (a) Perry, A.; Taylor, R. J. K. Chem. Commun. 2009, 3249.
(b) Klein, J. E. M. N.; Perry, A.; Pugh, D. S.; Taylor, R. J. K. Org. Lett.
2010, 12, 3446. (c) Moody, C. L.; Franckevici
Klein, J. E. M. N.; Taylor, R. J. K. Tetrahedron Lett. 2012, 53, 1897.
(11) (a) Jia, Y.-X.; Kundig, E. P. Angew. Chem., Int. Ed. 2009, 48,
̌
us, V.; Drouhin, P.;
̈
1636. (b) Dey, C.; Larionov, E.; Kundig, E. P. Org. Biomol. Chem.
2013, 11, 6734.
̈
(12) (a) Dey, C.; Kundig, E. P. Chem. Commun. 2012, 3064.
̈
(b) Hurst, T. E.; Gorman, R. M.; Drouhin, P.; Perry, A.; Taylor, R. J.
K. Chem.Eur. J. 2014, DOI: 10.1002./chem.201403917.
(13) X-ray data for 4a have been deposited with the Cambridge
Crystallographic Data Centre (CCDC 1013303), which can be
obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif.
(14) X-ray data for 5a have been deposited with the Cambridge
Crystallographic Data Centre (CCDC 1004040), which can be
obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif.
(15) The cyclopentanone precursors were prepared using the
procedure described for 4a; the 6- and 7-membered ring precursors
4g−j were synthesized from cyclohexanone or -heptanone via double
carboxylation with methyl magnesiumcarbonate: Stiles, M. J. Am.
Chem. Soc. 1959, 81, 2598. This was followed by double Mukaiyama
amide coupling (see Supporting Information for details).
(16) The previous conditions (Cu(OAc)2·H2O, KOtBu in DMF at
110 °C) were tried; however, the yield of the product was inferior
(∼54%). This drop in yield is mainly due to formation of oxidative
cleavage byproducts, namely N-methyl-3-alkyl-3-hydroxyoxindoles.
Formation of these byproducts is minimized under the Cu(OAc)2·
H2O/toluene conditions.
(17) X-ray data for cis- and trans-10 have been deposited with the
Cambridge Crystallographic Data Centre (CCDC 1004041 and
4903
dx.doi.org/10.1021/ol5024129 | Org. Lett. 2014, 16, 4900−4903