Reversal of Selectivity in C3-Allylation and Formal [3 + 2]-Cycloaddition of Spiro-epoxyoxindole: Unified Synthesis of Spiro-furanooxindole, (±)-N-Methylcoerulescine, (±)-Physovenine, and 3a-Allylhexahydropyrrolo[2,3-b]indole

Article abstract of DOI:10.1021/acs.orglett.7b00420

An effective Lewis acid catalyzed regioselective C3-allylation and a formal [3 + 2]-annulation reaction of spiro-epoxyoxindoles have been developed and can be accessed simply by changing the reaction conditions. This method has been successfully employed for the synthesis of spiro(pyrrolidinyloxindole), 3a-allylhexahydropyrrolo[2,3-b]indole, and furanoindoline.

Full text of DOI:10.1021/acs.orglett.7b00420

Letter
pubs.acs.org/OrgLett
Reversal of Selectivity in C3-Allylation and Formal [3 + 2]-
Cycloaddition of Spiro-epoxyoxindole: Unified Synthesis of Spiro-
furanooxindole, ( )‑N‑Methylcoerulescine, ( )-Physovenine, and 3a-
Allylhexahydropyrrolo[2,3‑b]indole
Saumen Hajra,*,† Sayan Roy,† and Subrata Maity†,‡
†Centre of Biomedical Research, Sanjay Gandhi Post-Graduate Institute of Medical Sciences Campus, Raebareli Road, Lucknow
226014, India
‡Department of Chemistry, Indian Institute of Technology Kharagpur, Kharagpur 721302, India
S
* Supporting Information
ABSTRACT: An effective Lewis acid catalyzed regioselective C3-allylation and a formal [3 + 2]-annulation reaction of spiro-
epoxyoxindoles have been developed and can be accessed simply by changing the reaction conditions. This method has been
successfully employed for the synthesis of spiro(pyrrolidinyloxindole), 3a-allylhexahydropyrrolo[2,3-b]indole, and furanoindo-
line.
he construction of small molecules bearing an all-carbon
Tquaternary center at the C3-position of oxindole has
garnered considerable research interest in recent years because of
the remarkable prevalence of such molecules in innumerable
natural products.1 The synthesis of spiro(pyrrolidinyloxindole)
alkaloids, namely, horsfiline2 and coerulescine,3 has attracted the
attention of synthetic chemists due to their intriguing molecular
architecture and their broad range of pharmacological properties,
which include anticancer, antimigraine, and contraceptive
activities. On the other hand, hexahydropyrrolo[2,3-b]indole
(HPI) ring systems with a carbon atom at the pseudobenzylic site
of oxindole are found in a wide range of biologically active indole
alkaloids4 such as physostigmine. In particular, HPI rings with
C3-3,3-dimethyl allyl substituents are present in natural products
such as flustramine B, the pseudophrynamines, and mollenine A.
Figure 1. Several representative spiro(pyrrolidinyloxindole) and
hexahydropyrrolo[2,3-b]indole alkaloids.
Analogously, the C3-1,1-dimethyl allyl motif is also present in
indole alkaloids such as brevicompanine B and flustramines A
and C. All of these molecules5 feature 3a-allylhexahydropyrrolo-
[2,3-b]indole as a subunit (Figure 1).
rearrangement by Kozlowski,6e (iv) double C−H activation for
The paramount clinical significance and the unique structural
the synthesis of 3,3′-disubstituted oxindoles by Taylor et al.,6f (v)
array of these molecules have inspired us to find a unified strategy
Cu-catalyzed intramolecular arylation-alkylation for the synthesis
of HPIs by Zhang,6g (vi) thiourea-catalyzed aldol reactions with
paraformaldehyde for the synthesis of HPIs and spiro-
(pyrrolidinyloxindole) by the Bisai group,6h and (vii) the
combination of Ru-catalyzed C−H functionalization with Pd-
to synthesize them or their congeners from a common precursor.
A number of synthetic strategies toward the construction of an
all-carbon quaternary center, followed by the synthesis of the
aforementioned molecules, have been reported thus far in the
literature.6 The examples of (i) the Pd6b- and Mo6c-catalyzed
allylic alkylation by Trost et al., (ii) catalytic allylation of 3-
aryloxindoles by merging Pd-catalysis and H-bond catalysis by
the Xiao group,6d (iii) catalytic Meerwein−Eschenmoser Claisen
Received: February 22, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00420
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
catalyzed allylic alkylation by Lautens et al.6i are worthy of
attention.
We therefore commenced our investigation on the regiose-
lective ring opening of model substrate N-methyl spiro-
epoxyoxindole 1a with allyltrimethylsilane by varying the Lewis
acid and solvent (Table 1; for details, see the SI). As previously
Scrutinizing all of these literature reports, one can recognize
C3-allyl oxindole−methanol or the corresponding ester as a
versatile antecedent. To the best of our knowledge, literature
reports of the Hosomi−Sakurai allylation of epoxides7 are sparse;
moreover, we know of no reports on the allylation of spiro-
epoxyoxindole. The result of the regioselective ring opening of
spiro-epoxyoxindole with nucleophiles such as indoles and
arenes by our group8 and later by Wei et al.9 encouraged us to
explore the regioselective allylation of spiro-epoxyoxindole with a
potent nucleophile such as allyltrimethylsilane. When the
reaction was performed using our previously developed method,8
only a small amount of the desired 3a-allyloxindole−methanol
product was obtained; however, a substantial amount of an
unexpected [3 + 2]-annulated product was obtained. The
existing literature confirmed the identity of this cycloadduct.11
Interestingly, this annulated spirofuranooxindole moiety is also
an important framework because it is present as a fundamental
structural unit in XEN907 and XEN402, which are in phase IIb
clinical trials for the treatment of chronic pain.10 Therefore, the
development of a method for the synthesis of both functionalities
with appreciative selectivity is highly desirable. Herein, we
describe a solvent-dependent, Lewis acid catalyzed, and highly
regioselective Hosomi−Sakurai allylation as well as a formal [3 +
2]-cycloaddition of spiro-epoxyoxindole (Scheme 1). This
a
Table 1. Optimization of Reaction Conditions
b
yield (%)
entry
Lewis acid
solvent
time (h)
2a
3a
1
Sc(OTf)3
In(OTf)3
Cu(OTf)2
BF3·OEt2
Y(OTf)3
Sc(OTf)3
Y(OTf)3
Y(OTf)3
Sc(OTf)3
In(OTf)3
Cu(OTf)2
Y(OTf)3
BF3·OEt2
DCE
1
22
20
23
18
<5
21
10
9
65
56
55
63
49
56
45
34
<5
10
15
<5
10
2
DCE
1
3
DCE
1.5
1
4
DCE
5
DCE
1.5
1
6
CH2Cl2
CH2Cl2
CHCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
7
6
8
6
9
1
78
72
70
24
26
10
11
12
13
1
1.5
6
6
a
Scheme 1. Reaction of Spiro-epoxyoxindole 1 with Allylsilane
N-Methyl spiro-epoxyoxindole 1a (0.57 mmol), trimethylallylsilane
(1.7 mmol), and Lewis acid (10 mol %) in solvent (4 mL) were stirred
b
at 0 °C. Isolated yield.
described, when the initial attempt was performed in DCE at 0
°C using 10 mol % of Sc(OTf)3, only 22% of 3-allyl-3-
(hydroxymethyl)oxindole 2a was obtained along with 65% of [3
+ 2]-cycloadduct 3a within 1 h (entry 1). Changing the Lewis
acid to In(OTf)3, Cu(OTf)2, Bi(OTf)3, Yb(OTf)3, Zn(OTf)2,
and BF3·OEt2 in DCE and CH2Cl2 did not improve the product
distribution. However, the cycloadduct 3a was obtained almost
exclusively using 10 mol % of Y(OTf)3 within 1.5 h, but the yield
was not satisfactory since a significant amount of undesired as
well as uncharacterized products was formed (entry 5). With this
promising result, we further applied Y(OTf)3 in other
chlorinated solvents like CH2Cl2 and CHCl3 but ended up
with unsatisfactory results (entry 7 and 8).
method also provides direct access to 3-allyl-3-(hydroxymethyl)-
oxindole, which was effectively utilized in the synthesis of
spiro(pyrrolidinyloxindole), 3a-allylhexahydropyrrolo[2,3-b]-
indole, and furanoindoline.
If we deeply explore a plausible mechanism for the Hosomi−
Sakurai allylation and the [3 + 2]-annulation reaction, the result
can be rationalized through the addition of allylsilane to reactive
2H-indol-2-one intermediate 4, formed upon treatment of spiro-
epoxyoxindole with a Lewis acid (Scheme 2). Path a in Scheme 2
However, to our delight, after the solvent was changed from
DCE to the more polar acetonitrile, we obtained the C3-allylated
product 2a almost exclusively in the presence of 10 mol % of
Sc(OTf)3 (entry 9). With the aforementioned encouraging result
in hand, we screened different Lewis acids in acetonitrile.
Whereas In(OTf)3, Cu(OTf)2, and Bi(OTf)3 displayed com-
parable results, BF3·OEt2, Yb(OTf)3, Y(OTf)3, and Zn(OTf)2
were not as successful. The reaction in THF with Sc(OTf)3 was
sluggish, whereas almost no reaction was observed in toluene.
Thus, DCE was shown to be the best solvent for annulation,
whereas the Hosomi−Sakurai allylation reaction was favored in
acetonitrile.
Scheme 2. Plausible Mechanism for C3-Allylation and [3 + 2]-
Annulation Reaction
shows the allylation reaction proceeding through the direct
addition of the allyl group to indolone intermediate 4 as a
consequence of C−Si bond cleavage; by contrast, path b leads to
the formation of a spiro-oxindole type annulated product, which
can be explained by the well-known β-silicon effect. Con-
sequently, both the solvent and the Lewis acid may strongly affect
the course of the reaction.
On the basis of the aforementioned optimization, we further
explored the substrate scope of the annulation and that of the
allylation reaction with an array of spiro-epoxyoxindoles and
allyltrimethylsilane. First, we executed the regioselective
Hosomi−Sakurai allylation reaction with spiro-epoxyoxindoles
bearing different protecting groups in acetonitrile (Figure 2). All
N-methyl, N-benzyl, N-allyl, and N-prenyl epoxides smoothly
B
DOI: 10.1021/acs.orglett.7b00420
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
N-benzyl spiro-epoxyoxindoles. However, a fine-tuning of the
Lewis acid to In(OTf)3 afforded almost exclusively the product of
the annulation reaction in good yield for unprotected spiro-
epoxyoxindoles. A number of these spirocyclic oxindoles were
obtained with a diastereomeric ratio of 1:1, although some of
them exhibited better selectivity. It was also noteworthy that the
cycloadduct 3a was partly converted to allylated product 2a when
treated with an excess of Sc(OTf)3 in acetonitrile at ambient
temperature, but the same was not observed in DCE. This result
validated our presumption that a polar solvent facilitated the C−
Si bond cleavage, whereas the β-silicon effect was predominant in
nonpolar solvent.
As previously mentioned, we were interested in 3-allyl-3-
(hydroxymethyl)oxindole moieties because they are key
precursors in the synthesis of a large variety of bioactive natural
products and pharmaceuticals. As an application of this reaction,
an initial effort was devoted to the synthesis of spiro-
(pyrrolidinyloxindole) (Scheme 3). This 3-allyl-3-(hydroxy-
Scheme 3. Synthesis of ( )-N-Methylcoerulescine,
( )-Physovenine, and 3a-Allylic Pyrroloindole from 2a
Figure 2. Substrate scope for the C3-allylation reaction.
underwent the allylation reaction with excellent regioselectivity
under the optimized conditions to give the corresponding
product in good to excellent yield (69−83%). Substituents on
the spiro-epoxyoxindoles had little effect on the rate of the
reactions. Whereas electron-donating substituents accelerated
the allylation reactions, causing them to be completed within 1−
2 h, electron-withdrawing substituents on the spiro-epoxyox-
indoles prolonged the reaction time to 4 h. The reaction was also
evaluated with unprotected spiro-epoxyoxindoles. Cu(OTf)2 in
acetonitrile was found to be the highest yielding catalyst,
flawlessly furnishing the allylation product 2r−t with excellent
selectivity.
Next, the scope of the [3 + 2]-annulation reaction employing
allyltrimethylsilane and different spiro-epoxyoxindoles was
investigated in DCE (Figure 3). Sc(OTf)3 (10 mol %) was
optimal for all N-methyl and N-allyl spiro-epoxyoxindoles,
whereas 10 mol % of BF3·OEt2was found to be the best catalyst in
terms of the yield and feasibility of the annulation reaction for all
methyl)oxindole 2a could be converted to ( )-N-methylcoer-
ulescine 8, applying a previously reported procedure.12 Again,
tosylation of the primary hydroxyl group in 3-allyl-3-
(hydroxymethyl)oxindole 2a followed by ozonolysis provided
another relevant intermediate, aldehyde 7. Furthermore,
aldehyde 7 was efficiently transformed into furanoindole when
treated with LiAlH4 at 0 °C. Expulsion of the −OTs group with
the assistance of superhydride (LiEt3BH) provided another
advanced intermediate 9 in satisfactory yield that could be easily
converted into the furanoindoline alkaloid ( )-physovenine via a
reported literature procedure.13 The synthetic potential and
effectiveness of this method were further elaborated through the
synthesis of 3a-allylhexahydropyrrolo[2,3-b]indole. Cyanation of
tosylated compound 6 by vigilant use of potassium cyanide
afforded oxindole cyanide. Partial hydrolysis of this crude cyanide
compound using hydrogen peroxide with sodium hydroxide in
methanol furnished amide 10 in good yield. This reaction was
followed by reductive cyclization accomplished by applying
LiAlH4 in boiling THF to furnish C3-allyl pyrroloindole. Boc
Figure 3. Substrate scope for the [3 + 2]-annulation reaction (10−22%
of the C3-allylation product was obtained in the case of 3a−l; 3m−o
were obtained almost exclusively (>95%)).
C
DOI: 10.1021/acs.orglett.7b00420
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
protection of this compound simplified its purification. This Boc-
protected 3a-allylhexahydropyrrolo[2,3-b] indole 11 could be
converted into the corresponding aldehyde compound, a
formidable intermediate for the synthesis of the pseudophryn-
amine family of 3a-prenylated hexahydropyrrolo[2,3-b]indole
alkaloids through the application of a previously reported
procedure.6h
Finally, we extended our strategy toward the asymmetric ring-
opening reaction of spiro-epoxyoxindole using several chiral
ligands but hardly found any encouraging results (for details, see
the SI). Only the chiral ligand L with Sc(OTf)3 furnished the
product 2a with very low enantioselectivity (5% ee, Scheme 4).
DEDICATION
■
This work is dedicated to Prof. Dipakranjan Mal (IIT Kharagpur)
on the occasion of his 65th birthday.
REFERENCES
■
(1) (a) Corey, E. J.; Guzman-Perez, A. Angew. Chem., Int. Ed. 1998, 37,
388. (b) Anthoni, U.; Christophersen, C.; Nielsen, P. H. In Alkaloids:
Chemical and Biological Perspectives; Pelletier, S. W., Ed.; Pergamon:
London, 1999; Vol. 13, pp 163−236. (c) Hibino, S.; Choshi, T. Nat.
Prod. Rep. 2001, 18, 66. (d) Jensen, B. S. CNS Drug Rev. 2002, 8, 353.
(e) Mermerian, A. H.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 5604.
(f) Ruiz- Sanchis, P.; Savina, S. A.; Albericio, F.; Alvarez, M. Chem. - Eur.
J. 2011, 17, 1388. (g) Dalpozzo, R.; Bartoli, G.; Bencivenni, G. Chem.
Soc. Rev. 2012, 41, 7247. (h) Mei, T. − S.; Patel, H. H.; Sigman, M. S.
Nature 2014, 508, 340. (i) Quasdorf, K. W.; Overman, L. E. Nature
2014, 516, 181.
Scheme 4. Asymmetric Version of C3-Allylation Reaction
(2) (a) Jossang, A.; Jossang, P.; Hadi, H. A.; Sevenet, T.; Bodo, B. J. Org.
Chem. 1991, 56, 6527. (b) For a review on the synthesis of
spirooxindole alkaloids, see: Marti, C.; Carreira, E. M. Eur. J. Org.
Chem. 2003, 2003, 2209.
(3) Anderton, N.; Cockrum, P. A.; Colegate, S. M.; Edgar, J. A.; Flower,
K.; Vit, I.; Willing, R. I. Phytochemistry 1998, 48, 437.
(4) (a) Triggle, D. J.; Mitchell, J. M.; Filler, R. CNS Drug Rev. 1998, 4,
87. (b) Santos, P. E.; Almeida, P. S.; Lobo, A. M.; Prabhakar, S.
Heterocycles 2001, 55, 1029. (c) Sorbera, L. A.; Castaner, J. Drugs Future
2003, 28, 18. (d) Crich, D.; Banerjee, A. Acc. Chem. Res. 2007, 40, 151.
(e) Pandey, G.; Khamrai, J.; Mishra, A. Org. Lett. 2015, 17, 952.
(5) (a) Daly, J. W.; Garraffo, H. M.; Pannell, L. K.; Spande, T. F. J. Nat.
In conclusion, we have accomplished a convenient, versatile,
Lewis acid-catalyzed and highly selective C3-allylation as well as a
formal [3 + 2]-annulation reaction of spiro-epoxyoxindoles that
can be accessed simply by manipulating the reaction conditions.
Whereas the annulation reactions furnished silicon-containing
spiro-furanooxindoles, the Hosomi−Sakurai allylation of the
spiro-epoxyoxindoles provided facile access to 3-allyl-3-
(hydroxymethyl)oxindole, a useful intermediate for the synthesis
of ( )-N-methylcoerulescine and the formal synthesis of
( )-physovenine. As a related application, 3-allyl-3-
(hydroxymethyl)oxindole 2a has been neatly transformed into
3a-allylhexahydropyrrolo[2,3-b]indole, a subunit of a large
number of members of the HPI-alkaloid family. Further studies
on the utility of spiro-epoxyoxindoles and their applications,
including asymmetric ring-opening reactions, are currently being
investigated in our laboratory.
Prod. 1990, 53, 407. (b) Peters, L.; Konig, G. M.; Terlau, H.; Wright, A.
̈
D. J. Nat. Prod. 2002, 65, 1633. (c) Smith, B. P.; Tyler, M. J.; Kaneko, T.;
Garraffo, H. M.; Spande, T. F.; Daly, J. W. J. Nat. Prod. 2002, 65, 439.
(d) Sprogøe, K.; Manniche, S.; Larsen, T. O.; Christophersen, C.
Tetrahedron 2005, 61, 8718. (e) Rochfort, S. J.; Moore, S.; Craft, C.;
Martin, N. H.; Van Wagoner, R. M.; Wright, J. L. C. J. Nat. Prod. 2009,
72, 1773.
(6) (a) Kawasaki, T.; Ogawa, A.; Terashima, R.; Saheki, T.; Ban, N.;
Sekiguchi, H.; Sakaguchi, K.; Sakamoto, M. J. Org. Chem. 2005, 70, 2957.
(b) Trost, B. M.; Frederiksen, M. U. Angew. Chem., Int. Ed. 2005, 44,
308. (c) Trost, B. M.; Zhang, Y. Chem. - Eur. J. 2010, 16, 296.
(d) Boucherif, A.; Duan, S.-W.; Yuan, Z.-G.; Lu, L.-Q.; Xiao, W.-J. Adv.
Synth. Catal. 2016, 358, 2594. (e) Linton, E. C.; Kozlowski, M. C. J. Am.
Chem. Soc. 2008, 130, 16162. (f) Klein, J. E. M. N.; Perry, A.; Pugh, D. S.;
Taylor, R. J. K. Org. Lett. 2010, 12, 3446. (g) Zhou, Y.; Xi, Y.; Zhao, J.;
Sheng, X.; Zhang, S.; Zhang, H. Org. Lett. 2012, 14, 3116. (h) De, S.;
Das, M. K.; Bhunia, S.; Bisai, A. Org. Lett. 2015, 17, 5922. (i) Yamamoto,
K.; Qureshi, Z.; Tsoung, J.; Pisella, G.; Lautens, M. Org. Lett. 2016, 18,
4954.
(7) (a) Hosomi, A. Acc. Chem. Res. 1988, 21, 200. (b) Naruta, Y.;
Maruyama, K. Chem. Lett. 1987, 16, 963. (c) Imai, T.; Nishida, S. J. Org.
Chem. 1990, 55, 4849.
(8) (a) Hajra, S.; Maity, S.; Maity, R. Org. Lett. 2015, 17, 3430.
(b) Hajra, S.; Maity, S.; Roy, S. Adv. Synth. Catal. 2016, 358, 2300.
(9) Luo, M.; Yuan, R.; Liu, X.; Yu, L.; Wei, W. Chem. - Eur. J. 2016, 22,
9797.
(10) Chafeev, M.; Chowdhury, S.; Fraser, L.; Fu, J. − M.; Langille, J.;
Liu, S.-F.; Sun, J.-Y.; Sun, S.-Y.; Sviridov, S.; Wood, M.; Zenova, A.; Jia,
Q.; Cadieux, J.-J.; Gauthier, S. J.; Douglas, A. F.; Hsieh, T.; Chakka, N.;
Cikojevic, Z. PCT Int. Appl. WO2010045251A2 20100422, 2010.
(11) Sugita, Y.; Kimura, Y.; Yokoe, I. Tetrahedron Lett. 1999, 40, 5877.
(12) Kim, M.; Kim, G. Bull. Korean Chem. Soc. 2012, 33, 1821.
(13) Kulkarni, M. J.; Dhondge, A. P.; Borhade, A. S.; Gaikwad, D. D.;
Chavhan, S. W.; Shaikh, Y. B.; Nigdale, V. B.; Desai, M. P.; Birhade, D.
R.; Shinde, M. P. Eur. J. Org. Chem. 2009, 2009, 3875.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b00420.
Experimental details, characterization data and spectra for
all new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: saumen.hajra@cbmr.res.in.
ORCID
Saumen Hajra: 0000-0003-0303-4647
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank SERB, New Delhi (EMR/2016/001161), for
providing financial support. S.R. and S.M. thank CSIR, New
Delhi, for their fellowships. We thank the Director of CBMR for
providing research facilities.
D
DOI: 10.1021/acs.orglett.7b00420
Org. Lett. XXXX, XXX, XXX−XXX