Rapid Assembly of Functionalised Spirocyclic Indolines by Palladium-Catalysed Dearomatising Diallylation of Indoles with Allyl Acetate

Article abstract of DOI:10.1002/chem.201403940

Herein, we report the application of allyl acetate to the palladium-catalysed dearomatising diallylation of indoles. The reaction can be carried out by using a readily available palladium catalyst at room temperature, and can be applied to a wide range of substituted indoles to provide access to the corresponding 3,3-diallylindolinines. These compounds are versatile synthetic intermediates that readily undergo Ugi reactions or proline-catalysed asymmetric Mannich reactions. Alternatively, acylation of the 3,3-diallylindolinines with an acid chloride or a chloroformate, followed by treatment with aluminium chloride, enables 2,3-diallylindoles to be prepared. By using ring-closing metathesis, functionalised spirocyclic indoline scaffolds can be accessed from the Ugi products, and a dihydrocarbazole can be prepared from the corresponding 2,3-diallylindole.

Full text of DOI:10.1002/chem.201403940

DOI: 10.1002/chem.201403940
Full Paper
&
Synthetic Methods
Rapid Assembly of Functionalised Spirocyclic Indolines by
Palladium-Catalysed Dearomatising Diallylation of Indoles with
Allyl Acetate
[
a]
Persis Dhankher, Laure Benhamou, and Tom D. Sheppard*
Abstract: Herein, we report the application of allyl acetate
to the palladium-catalysed dearomatising diallylation of in-
doles. The reaction can be carried out by using a readily
available palladium catalyst at room temperature, and can
be applied to a wide range of substituted indoles to provide
access to the corresponding 3,3-diallylindolinines. These
compounds are versatile synthetic intermediates that readily
undergo Ugi reactions or proline-catalysed asymmetric Man-
nich reactions. Alternatively, acylation of the 3,3-diallylindoli-
nines with an acid chloride or a chloroformate, followed by
treatment with aluminium chloride, enables 2,3-diallylindoles
to be prepared. By using ring-closing metathesis, functional-
ised spirocyclic indoline scaffolds can be accessed from the
Ugi products, and a dihydrocarbazole can be prepared from
the corresponding 2,3-diallylindole.
Introduction
readily available indoles into polyfunctionalised spirocyclic in-
[9]
dolines, which enables the incorporation of functional groups
at a variety of different positions in the scaffold.
Medicinal chemistry has traditionally focused on the synthesis
of aromatic heterocycles and related derivatives as lead com-
pounds due to their drug-like physical properties and synthetic
accessibility. However, it is widely considered that an increasing
focus on three-dimensional structures, which incorporate
We envisaged that introduction of two allyl groups at the 3
position of an indole core, with concomitant dearomatisation,
would enable the formation of a diallylindolinine. Such a com-
pound is potentially a very versatile synthetic intermediate
that can undergo reactions with a variety of nucleophiles and
electrophiles at the imine moiety, cross-coupling/CÀH function-
alisation reactions on the aromatic ring, and can be readily
converted into a spirocyclic alkaloid-like structure by ring-clos-
ing metathesis (Scheme 1). Given the large number of indoles
3
[1]
a greater proportion of sp carbons, is probably desirable, in
order for medicinal chemistry programmes to be successful
against more complex drug targets. Whilst natural products
can provide suitable three-dimensional functionalised architec-
tures, such compounds are often highly complex, difficult to
synthesise, and display undesirable physical properties. As
a consequence, there is considerable interest in the develop-
ment of synthetic routes to access small chiral saturated rings
[
2]
[3–4]
[5]
including cyclopropanes, oxetanes
and azetidines, as
[6]
well as benzofused heterocyclic systems, such as indolines
[
7]
and dihydrobenzofurans. In addition, the incorporation of
these motifs into spirocyclic frameworks can lead to increased
molecular complexity whilst maintaining a relatively low mo-
[
4,8]
lecular weight. Such systems can potentially provide a struc-
turally rigid three-dimensional core, which can be functional-
ised at several sites to provide a library of drug-like com-
pounds. Herein, we report a short synthetic route to convert
Scheme 1. Proposed synthetic route to spirocyclic indolines (RCM=ring-
closing metathesis).
that are readily available commercially, a diverse range of func-
tionalised spirocylic scaffolds could readily be constructed. The
spirocyclic ring structure embedded in these scaffolds has not
[
a] Dr. P. Dhankher, Dr. L. Benhamou, Dr. T. D. Sheppard
Department of Chemistry, University College London
[10]
2
0 Gordon St, London, WC1H 0AJ (UK)
been widely explored in existing drugs, though it forms part
of the polycyclic frameworks of the ajmaline alkaloid natural-
E-mail: tom.sheppard@ucl.ac.uk
[11]
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403940.
product family, which contains compounds possessing anti-
[12]
[13]
arrhythmic
and antiplasmodial activity,
some of which
ꢀ
2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
have found application in medical treatment.
This is an open access article under the terms of the Creative Commons At-
tribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.
Our synthetic plan involved the development of a Pd-cata-
lysed allylation procedure to achieve regioselective introduc-
Chem. Eur. J. 2014, 20, 13375 – 13381
13375 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion of the allyl groups in a dearomatising reaction, to gener-
ate the desired isoindolinine. Whilst there is good precedent
for the Pd-catalysed allylation of indoles with a range of allyl
(entry 2), though large quantities of allyl acetate appeared to
significantly slow down the reaction (entry 3). In the absence
of base, no conversion to either 3a or 4a was seen (entry 4).
By increasing the quantity of base in the reaction to three
equivalents, and by employing five equivalents of allyl acetate,
almost complete conversion of indole into 3a and 4a was ob-
served with 3a becoming the major product (entry 5). The
choice of ligand proved to be critical to controlling both the
reactivity and the product distribution. A range of bidentate
phosphines were explored, with ethylenebis(diphenylphos-
phine) (dppe) proving to be very ineffective (entry 6) and both
(2,2’-bis(diphenylphosphino)-1,1’-binaphthyl) (BINAP) and 4,5-
bis(diphenylphosphino)-9,9-dimethylxanthene (Xantphos) lead-
ing to lower conversions and selectivity in comparison to dppf
(entries 7 and 8). However, with bis-[2-(diphenylphosphino)-
phenyl]ether (DPEPhos), we were able to obtain an excellent
conversion to the desired diallylindolinine 3a with very high
selectivity (entry 9). Pleasingly, an 82% isolated yield of 3a
could be obtained with only 5 mol% palladium catalyst at
room temperature.
[
14–18]
sources,
and even for the dearomatising allylation of 3-
substituted indoles with allyl carbonates, with allyl alcohols in
[
15]
combination with organoboranes or by rearrangement of N-
[
14,19–22]
alloc protected indoles,
the direct use of allyl acetate in
[
23]
such reactions has proved challenging to date.
Although
highly activated allylic esters containing two conjugated aro-
matic rings can be successfully used in Pd-catalysed allylation
[
24]
reactions, there is only a single report of the Pd-catalysed re-
action of indole 1a with allyl acetate 2, and the reaction was
reported to produce a relatively complex mixture of N and C
allylated products 3a–6a from which 3-allylindole 4a was iso-
[
23]
lated in up to 54% yield.
Allyl acetate is a considerably
cheaper starting material than the carbonates and carbamates
typically employed, because it is a bulk chemical that can be
prepared on an industrial scale by direct acetoxylation of pro-
[
25]
pene. Therefore, we decided to explore whether it could be
successfully employed in the direct Pd-catalysed diallylation of
indole.
With practical conditions in hand, we went on to explore
the scope of this dearomatising double allylation reaction
(
Scheme 3). A wide range of substituted indoles 1a–o were
Results and Discussion
An initial test reaction of indole with Pd catalyst, 1,1’-bis(diphe-
nylphosphino)ferrocene (dppf) ligand, allyl acetate 2 and K CO3
2
led to the formation of a mixture of the desired product 3a, 3-
allylindole 4a, as well as traces of 1,3-diallylindole 6a
(Scheme 2; Table 1, entry 1). By increasing the number of
equivalents of allyl acetate, a higher conversion was observed
Scheme 2. Screening of reaction conditions for the diallylindolinine forma-
tion.
Table 1. Screening of reaction conditions.
Entry
K
2
CO
3
[equiv] 2 [equiv] Ligand (5 mol%) Conv. [%]
Ratio
[
a]
(yield [%] 3a) 3a/4a
[
[
[
[
[
[
[
[
[
b]
b]
b]
b]
c]
c]
c]
c]
c]
1
2
3
4
5
6
7
8
9
2
2
2
0
3
3
3
3
3
1.5
2.2
7
5
5
5
5
5
5
dppf
dppf
dppf
dppf
dppf
dppe
rac-BINAP
Xantphos
DPEPhos
53 (15)
62
1:2
1:1
1:5
–
2:1
–
29
0
>95
13
63 (45)
76 (47)
[d]
[d]
1:1
5:3
>10:1
>95 (82)
1
[
a] Determined by H NMR. [b] Reaction was performed at 408C. [c] Reac-
Scheme 3. Scope of the Pd-catalysed diallylation reaction. [a]Reaction per-
formed at 508C for 8–24 h.
tion was performed at RT. [d] Small quantities of 6a were also observed.
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13376 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
converted to the corresponding diallylindolenines 3a–o in
generally good to excellent yield. Notably, the reaction was tol-
erant of alkoxy or alkyl substituents at any position on the
benzene ring (3b–g). However, the allylation reactions of sev-
eral haloindoles were somewhat sluggish at room temperature,
requiring heating at 508C in order to obtain reasonable yields
compounds, if any, are plausible reaction intermediates. Nei-
ther 5a nor 6a apparently underwent rearrangement under
the reaction conditions, but small quantities of 6a were pro-
duced from 5a. This indicates that 5a and 6a are not plausible
intermediates in the formation of diallylindolinine 3a. However,
3-allylindole 4a was completely converted into a mixture of
3a and 6a upon resubmission to the reaction conditions. The
diallylindolinine 3a did not undergo conversion into 6a upon
resubmission to the reaction conditions, but some degradation
of 3a did take place suggesting that prolonged exposure of
the product to the reaction conditions will have a detrimental
effect on the reaction yield. These observations suggest that
only 4a is an intermediate in the reaction, and that the forma-
tion of by-products 5a and 6a is solely a result of competing
N-allylation on reaction of 1a or 4a with the electrophilic p-
allyl complex. The use of K CO as base in related reactions has
[
26]
of the desired product (3h-.j). Pleasingly, the presence of
a substituent at C-2 did not impair the reaction, despite the
potential steric crowding around the reaction site (3l–o). The
presence of a tert-butyldimethylsilyl(TBS)-protected alcohol at
C-2 was compatible with the reaction, giving a good yield of
the diallylindolinine 3o considering the presence of this rela-
tively large substituent so close to the newly formed quaterna-
ry carbon centre. As was anticipated, electron-rich indoles
were generally better substrates for the reaction (3b, d, f, m).
In contrast, highly electron-deficient systems did not form the
diallylindolinine at all, with the N-allylindoles 7a and 7b being
obtained as the major product from reactions of 5-nitroindole
and 2-methyl-5-nitroindole, respectively. No allylation or dially-
lation of N-methylindole was observed under these conditions,
demonstrating that a free NH is essential for the reaction to
take place.
2
3
been reported to promote N-allylation over C-allylation due to
[24]
the formation of a looser ion pair, but this is obviously not
a significant factor in our reaction. It should be noted that the
use of lithium or sodium carbonates in this diallylation reaction
was ineffective.
With these useful diallylindolinine building blocks in hand,
we then began to investigate the reactivity of these com-
pounds in further transformations (see Scheme 5). To rapidly
Given the difficulties initially encountered with achieving se-
lectivity in the indole allylation reaction, we recognised that
several different reaction pathways may be operative
(Scheme 4). A likely possibility is that the reaction proceeds di-
rectly through allylation at C-3 to give 4a, followed by
a second allylation at C-3 to give the diallylindolinine 3a. How-
ever, it was also possible that 4a might undergo N-allylation to
give 6a, followed by a (potentially Pd-catalysed) rearrange-
ment to generate diallylindolinine 3a. Conversely, the by-prod-
uct 6a may be formed by rearrangement of 3a. In this latter
case, the yield of 3a would be eroded during prolonged expo-
sure to the reaction conditions as it was gradually converted
into the undesired by-product 6a. Furthermore, initial N-allyla-
tion of 1a to give 5a, could be followed by (potentially, Pd-
catalysed) rearrangement to generate 4a. Therefore, we syn-
[
27]
[27]
thesised pure samples of 4a, 5a and 6a and resubmitted
them to the reaction conditions to determine, which of these
Scheme 4. Identification of possible reaction pathways by resubmission of
potential intermediates to the reaction conditions. Conditions: [Pd(allyl)Cl]
2.5 mol%), 2 (5 equiv), DPEPhos (5 mol%), K CO (3 equiv), MeCN, 24h, RT.
2
Scheme 5. Ugi reactions of the diallylindolinines (d.r.=diastereomeric ratio).
[a]Reaction at 508C; [b]7 d reaction time.
(
2
3
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13377 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
introduce diverse functionality onto the indoline core, we ex-
plored the use of the diallylindolinines 3 in multicomponent
reactions. Multicomponent reactions offer a highly efficient
route to construct functionalised molecules based on a central
[
28]
core structure. The use of imines and their derivatives, in
combination with isonitriles and a range of different nucleo-
philes, has often proved to be a highly effective strategy for
constructing bioactive small molecules of potential interest.
Indeed, the classical Ugi reaction and closely related processes
have proved to be of great value in the synthesis of a vast
array of medicinally relevant structures containing an a-amino-
[
29]
amide core. We envisaged that the highly stable imine unit
present in the indolinine core should be an excellent building
block for use in Ugi reactions, and surprisingly, the use of such
compounds in these multicomponent reactions does not seem
to have previously been explored. The diallyl indolinines 3
proved to be excellent substrates for Ugi reactions under stan-
dard conditions (Scheme 5). Reaction of selected indolinines
with a diverse selection of carboxylic acids and isocyanides in
MeOH gave access to a wide range of multicomponent prod-
ucts 8a–n (Scheme 5). As well as simple aliphatic and aromatic
groups (8a–d), products containing heterocyclic rings (8e–g),
primary amides (8h), activated chlorides (8i) and carbamate-
protected amines (8j–k) could be prepared. Remarkably, even
unprotected a-aminoacids (8l–m) and a-hydroxyacids (8n)
could be used directly in the multicomponent reactions, al-
though, as with many other Ugi reactions using chiral compo-
Scheme 6. Acylation of diallylindolinines to give 2-hydroxyindolines and 2,3-
diallylindoles by subsequent rearrangement. [a]From purified 10a/10c; yield
for the rearrangement step.
[30]
nents, only low levels of diastereoselectivity were observed.
An alternative strategy for functionalisation of the indolinine
was devised by reaction of the imine unit with a chloroformate
[
31]
or acid chloride to give access to 2-chloroindolines, which
after hydrolysis during work-up gave the corresponding 2-hy-
droxyindolines 10a–c in good yield (Scheme 6). Alternatively,
acylation with an acid chloride, followed by quenching with
methanol could be used to access a 2-methoxyindoline 11 d.
Acylation of 2-methylindolinine 3l with methyl chloroformate
resulted in the formation of N-acylenamine 12 in good yield.
The 2-hydroxyindolines 10a and 10c readily underwent rear-
rangement to the corresponding 2,3-diallylindoles 13a and
Scheme 7. Asymmetric Mannich reactions of diallylindolinines (e.r.=enantio-
meric ratio). For compound 14c, the reaction was carried out in Acetone/
DMSO.
13c upon treatment with aluminium trichloride, providing
a convenient route to an alternative structural motif. The acyla-
tion and rearrangement reactions could be conveniently com-
bined into a single process to provide access to the 2,3-dially-
lindoles 13d–f without the need to isolate the intermediate 2-
hydroxyindolines.
[
33]
closing metathesis was studied (Scheme 8). A selection of
the Ugi products (8a–c, e, i–k, n) underwent efficient ring-clos-
ing metathesis by using Grubbs’ first-generation catalyst in di-
chloromethane as solvent. The corresponding spirocyclic indo-
lines 15 were isolated in good to excellent yield. The 2-hydrox-
yindolines 10a and b could also be smoothly converted into
the corresponding spirocyclic indolines 16 in good yield. A di-
hydrocarbazole 17d could also be accessed by a ring-closing
metathesis reaction of 2,3-diallylindole 13d.
A strategy for desymmetrising the achiral diallylindolines 3
through asymmetric addition of a nucleophile to the imine
group was also explored (Scheme 7). Gratifyingly, we found
that l-proline-catalysed Mannich reaction of acetone with
three diallylindolinines (3a, d, j) gave the corresponding 3-ami-
noketones 14a–c in good yield and with very high enantiose-
lectivity. During the preparation of this manuscript, similar con-
ditions were reported for the asymmetric Mannich reaction of
Conclusion
[
32]
closely related 3,3-disubstituted indolinines, so this reaction
was not explored in further detail.
We have described the application of the bulk chemical allyl
acetate to the palladium-catalysed dearomatising diallylation
of indoles. The choice of ligand was found to be critical to con-
With a range of different substituted indolines in hand, the
synthesis of the corresponding spirocyclic indolines via ring
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13378 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
access a variety of 2,3-diallylindoles via acylation and rear-
rangement.
Experimental Section
General methods
All chemicals were purchased from Sigma–Aldrich, Acros, Alfa
Aesar or Santa Cruz Biotechnology and used without further purifi-
cation. Samples of functionalised carboxylic acids for the Ugi reac-
tions were provided by GlaxoSmithKline. 1-Allyl-1H-indole (5a), 3-
allyl-1H-indole (4a), 1,3-diallyl-1H-indole (6a) were synthesized ac-
[27]
cording to literature procedures.
Anhydrous THF, dichlorome-
thane and acetonitrile were purchased from Fisher Scientific. All
other solvents were used as received. PE refers to petroleum ether.
Flash-column chromatography was carried out by using normal-
phase silica gel (33–70 mm) supplied by VWR. Thin-layer chroma-
tography was carried out by using Merck TLC Silica gel 60 F254
plates and products were visualized by using combinations of UV
light (l=254 nm) and potassium permanganate (KMnO ) when re-
4
quired.
General procedure A: Synthesis of 3,3-diallyl-3H-indolinines
(
3)
The indole (1 equiv), [Pd(allyl)Cl)]2 (2.5 mol%), DPEPhos (5 mol%)
and K CO (3 equiv) were placed in an oven-dried carousel tube.
2
3
À1
After three vacuum/Ar cycles, acetonitrile (Cꢀ0.025 molL ) and
allyl acetate (5 equiv) were successively added. The heterogeneous
mixture was stirred at RT for 18–24 h before addition of water. The
Scheme 8. Synthesis of spirocyclic indolines and a carboazole by ring-closing
metathesis.
solution was extracted with Et O and washed with water. The com-
2
bined organic layers were dried with Na SO , filtered, and volatiles
2
4
were removed under vacuum. Purification by flash chromatogra-
trolling the product distribution, and under the optimised con-
ditions, the reaction proceeds efficiently and selectively at
room temperature by using 5 mol% of a readily available palla-
dium catalyst. This procedure enables the rapid assembly of
a range of substituted indolines from readily available indoles
by short synthetic sequences, including a selection of com-
pounds containing the spirocyclic indoline motif. A diverse
array of functional groups can be introduced into the scaffold
through Ugi multicomponent reactions, and the achiral scaf-
fold can also be desymmetrised by an asymmetric Mannich re-
action. Many of these molecules incorporate hydrogen-bond
donor and acceptor groups and possess potentially useful
drug-like properties (Figure 1). Furthermore, the indolinines
generated from the diallylation reaction can also be used to
phy on SiO gave the corresponding 3,3-diallyl-3H-indole com-
2
pound 3.
General procedure B: Synthesis of Ugi products (8)
The carboxylic acid (1 equiv) and the isocyanide (1 equiv) were
added to a solution of 3,3-diallyl-3H-indole (1 equiv) in MeOH (ca.
À1
0
.25 molL ). The reaction mixture was stirred for 2–24 h at RT,
before evaporation of the volatiles under vacuum. Pure com-
pounds were obtained by washing the crude residue with PE, or
by purification by column chromatography on SiO .
2
General procedure C: Synthesis of 3,3-diallyl-2-hydroxyindo-
lines (10)
The chloroformate or acid chloride (1 equiv) was added to a solu-
À1
tion of the 3,3-diallyl-3H-indole in CH Cl (ca. 0.07 molL ), and the
2
2
reaction was left to stir for 30 min at RT, before addition of saturat-
ed NaHCO . After extraction with CH Cl (three times), the com-
3
2
2
bined organic layers were washed with water, dried over Na SO
2
4
and filtered through cotton wool. Pure compounds were obtained
by evaporation of the volatiles under reduced pressure or by purifi-
cation by column chromatography on SiO2.
General procedure D: Preparation of 2,3-diallylindoles (13)
from 2-hydroxy-3,3-diallylindolines (10)
Aluminium chloride (1.1 equiv) was added to a solution 3,3-diallyl-
À1
2
-hydroxyindoline (1.0 equiv) in CH Cl2 (ca. 1 molL ) at RT. The
2
Figure 1. Calculated properties of selected indolines.
mixture was stirred for 30 min before addition of NEt (2 equiv).
3
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13379 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
After 5 min at RT, water was added, and the product was extracted
with CH Cl (three times). The combined organic layers were dried
General procedure I: Synthesis of spirocyclic indolines (16)
2
2
First-generation Grubbs’ catalyst (15 mol%) was added to a de-
gassed solution of the corresponding Ugi product (1 equiv) in
over Na SO . After evaporation, the crude material was purified by
2
4
filtration through a small pad of SiO to give the rearranged prod-
2
À1
CH Cl (ca. 0.06 molL ) at 458C. The reaction mixture was heated
2
2
uct.
at reflux for 24 h under argon before evaporation of the solvent
under reduced pressure. The crude residue was purified by column
Preparation of 2,3-diallylindoles (13) from 3,3-diallylindoli-
nines (3)
chromatography on SiO to give the desired product.
2
General procedure J: Synthesis of spirocyclic indolines (17)
General procedure E: Using an acyl chloride
The corresponding substituted 3,3-diallyl-2-hydroxyindoline com-
The acyl chloride (1 equiv) was added to a solution of 3,3-diallyl-
À1
pound was added to a refluxing solution of first generation
3
H-indole (1 equiv) in CH Cl (ca. 0.3 molL ) at room temperature.
2
2
À1
Grubbs’ catalyst (15 mol%) in CH Cl (ca. 0.04 molL ) under argon.
After 30 min, the reaction was quenched by addition of water and
the mixture was extracted with CH Cl (three times). The combined
2
2
The reaction was heated at reflux for 24 h before evaporation of
the volatiles under vacuum. The crude material was purified by
2
2
organic layers were dried over Na SO and filtered. Evaporation of
2
4
column chromatography on SiO to yield the desired product.
the volatiles give the 3,3-diallyl-2-hydroxyindoline which was di-
2
À1
rectly dissolved in CH Cl (ca. 0.3 molL ) and AlCl (1.1 equiv) was
2
2
3
added. After 30 min at room temperature, the reaction was
General procedure K: Synthesis of dihydro-1H-carbazole (18)
quenched with saturated NaHCO3 before extraction with CH Cl2
2
Grubbs’ first-generation catalyst (5 mol%) was added to a dried
(
three times). The combined organic phases were washed with
and degassed solution of 2,3-diallyl-1H-indole in CH Cl . The mix-
2
2
water, dried over Na SO and filtered. After evaporation of the vola-
2
4
ture was heated overnight at 558C before evaporation of the vola-
tiles under reduced pressure. The residue obtained was purified by
column chromatography on SiO2.
tiles under vacuum, the crude residue was purified by a filtration
on a small pad of SiO using PE/Et O (90:10) as eluent.
2
2
General procedure F: Using a chloroformate
Acknowledgements
The chloroformate (1 equiv) was added to a solution of 3,3-diallyl-
À1
3
H-indole (1 equiv) in CH Cl (ca. 0.2 molL ) at room temperature.
2
2
After 30 min, the reaction was quenched by addition of saturated
NaHCO , and the mixture was extracted with CH Cl (three times).
We would like to acknowledge the UCL drug discovery PhD
programme (PhD studentship to PD), the Engineering and
Physical Sciences Research Council (EP/K001183/1, PDRA fund-
ing to L.B.) and GlaxoSmithKline (supply of selected functional-
ised carboxylic acids) for supporting this work.
3
2
2
The combined organic layers were washed with water and dried
over Na SO before filtration. Evaporation of the volatiles gave the
2
4
3
,3-diallyl-2-hydroxy-indoline derivative which was directly dis-
À1
solved in CH Cl (ca. 0.2 molL ) and AlCl (1.1 equiv) was added.
2
2
3
After 30 min at RT, NEt (2 equiv) was added. The solution was left
3
to stir for 5 min before addition of a saturated solution of K CO
Keywords: allylation · heterocycles · homogeneous catalysis ·
indoles · palladium
2
3
and extraction with CH Cl2 (three times). The combined organic
2
phases were washed with water, dried over Na SO and filtered.
2
4
After evaporation of the volatiles under vacuum, the crude residue
was purified by filtration through a small pad of SiO using PE/Et O
[
1] F. Lovering, J. Bikker, C. Humblet, J. Med. Chem. 2009, 52, 6752–6756.
2
2
[2] a) F. Gnad, O. Reiser, Chem. Rev. 2003, 103, 1603–1624; b) A. Gagnon,
M. Duplessis, L. Fader, Org. Prep. Proced. Int. 2010, 42, 1–69; c) A. Reich-
elt, S. F. Martin, Acc. Chem. Res. 2006, 39, 433–442; d) S. Ishikawa, T. D.
Sheppard, J. M. D’Oyley, A. Kamimura, W. B. Motherwell, Angew. Chem.
(
100:0 to 90:10) as eluent.
Asymmetric synthesis of Mannich reaction products (15)
2
013, 125, 10244–10247; Angew. Chem. Int. Ed. 2013, 52, 10060–10063.
[
3] a) J. A. Burkhard, G. Wuitschik, M. Rogers-Evans, K. Mꢁller, E. M. Carreira,
Angew. Chem. 2010, 122, 9236–9251; Angew. Chem. Int. Ed. 2010, 49,
General procedure G
9
052–9067; b) K. F. Morgan, I. A. Hollingsworth, J. A. Bull, Chem.
l-Proline (30 mol%) was added at 08C to a solution of 3,3-diallyl-
Commun. 2014, 50, 5203–5205.
3
0
H-indole (1 equiv) in a mixture of acetone/CHCl (4.5:1, ca.
.022 molL ). The reaction mixture was allowed to warm up
3
À1
[4] G. Wuitschik, M. Rogers-Evans, A. Buckl, M. Bernasconi, M. Mꢂrki, T.
Godel, H. Fischer, B. Wagner, I. Parrilla, F. Schuler, Angew. Chem. 2008,
slowly to RT and stirred for two days. Evaporation of the solvent
1
20, 4588–4591; Angew. Chem. Int. Ed. 2008, 47, 4512–4515.
followed by purification by column chromatography on SiO gave
2
[5] a) A. Brandi, S. Cicchi, F. M. Cordero, Chem. Rev. 2008, 108, 3988–4035;
b) F. G. West, T. M. Bott, Heterocycles 2012, 84, 223–264; c) B. Alcaide, P.
Almendros, C. Aragoncillo, Curr. Opin. Clin. Nutr. Metab. Care Curr. Opin.
Drug. Disc. 2010, 13, 685–697.
the Mannich product.
General procedure H
[
6] D. Liu, G. Zhao, L. Xiang, Eur. J. Org. Chem. 2010, 3975–3984.
l-Proline (30 mol%) was added at 08C to a solution of 3,3-diallyl-
[7] a) T. D. Sheppard, J. Chem. Res. 2011, 35, 377–385; b) F. Bertolini, M.
Pineschi, Org. Prep. Proced. Int. 2009, 41, 385–418.
3
0
H-indole (1 equiv) in a mixture of acetone/DMSO (4:1, ca.
.016 molL ). The solution was allowed to warm up slowly to RT
À1
[8] a) R. J. Duffy, K. A. Morris, D. Romo, Tetrahedron 2009, 65, 5879–5892;
b) J. T. Lowe, M. D. Lee, L. B. Akella, E. Davoine, E. J. Donckele, L. Durak,
J. R. Duvall, B. Gerard, E. B. Holson, A. Joliton, S. Kesavan, B. C. Lemercier,
H. Liu, J. C. Mariꢃ, G. Muncipinto, M. Welzel-O’Shea, L. M. Panko, A.
Rowley, B. C. Suh, M. Thomas, F. F. Wagner, J. Wei, M. A. Foley, L. A. Mar-
caurelle, J. Org. Chem. 2012, 77, 7187–7211; c) M. A. Perry, R. R. Hill,
S. D. Rychnovsky, Org. Lett. 2013, 15, 2226–2229.
and stirred for two days. The reaction mixture was diluted with di-
ethyl ether and washed with saturated NaHCO . The product was
3
extracted with Et O (three times), and the combined organic layers
2
were washed with water, brine and dried with MgSO4. After filtra-
tion and removal of the solvents under reduce pressure, the crude
product was purified by column chromatography on SiO2.
[9] M. Li, P. A. Woods, M. D. Smith, Chem. Sci. 2013, 4, 2907–2911.
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13380 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
10] R. D. Taylor, M. MacCoss, A. D. G. Lawson, J. Med. Chem. 2014, 57, 5845–
859.
[23] W. E. Billups, R. S. Erkes, L. E. Reed, Synth. Commun. 1980, 10, 147–154.
[24] a) M. Bandini, A. Melloni, A. Umani-Ronchi, Org. Lett. 2004, 6, 3199–
3202.
[25] H. A. Wittcoff, B. G. Reuben, J. S. Plotkin, Eds. Industrial Organic Chemi-
cals, John Wiley & Sons Inc, 2012.
[26] B. Alcaide, P. Almendros, A. Luna, S. Cembellꢄn, M. Arnꢅ, L. R. Domingo,
Chem. Eur. J. 2011, 17, 11559–11566.
[27] Y. Zhang, D. Stephens, G. Hernandez, R. Mendoza, O. V. Larionov, Chem.
Eur. J. 2012, 18, 16612–16615.
5
[
11] a) H. Arai, Y. Hirasawa, A. Rahman, I. Kusumawati, N. C. Zaini, S. Sato, C.
Aoyama, J. Takeo, H. Morita, Bioorg. Med. Chem. 2010, 18, 2152–2158;
b) J. Li, T. Wang, P. Yu, A. Peterson, R. Weber, D. Soerens, D. Grubisha, D.
Bennett, J. M. Cook, J. Am. Chem. Soc. 1999, 121, 6998–7010; c) X. Fu,
J. M. Cook, J. Am. Chem. Soc. 1992, 114, 6910–6912; d) S. A. Saeed, R. U.
Simjee, F. Mahmood, N. N. Rahman, J. Pharm. Pharmacol. 1993, 45, 715–
719.
[
12] S. G. Wanner, H. Glossmann, H. G. Knaus, R. Baker, W. Parsons, K. M. Rup-
precht, R. Brochu, C. J. Cohen, W. Schmalhofer, M. Smith, V. Warren,
M. L. Garcia, G. J. Kaczorowski, Biochemistry 1999, 38, 11137–11146.
13] Y. Hirasawa, H. Arai, K. Zaima, R. Oktarina, A. Rahman, W. Ekasari, A. Wi-
dyawaruyanti, G. Indrayanto, N. C. Zaini, H. Morita, J. Nat. Prod. 2009,
[28] a) A. Dçmling, I. Ugi, Angew. Chem. 2000, 112, 3300–3344; Angew.
Chem. Int. Ed. 2000, 39, 3168–3210; b) L. El Kaim, L. Grimaud, Tetrahe-
dron 2009, 65, 2153–2171.
[29] a) L. Banfi, R. Riva, A. Basso, Synlett 2010, 1, 23–41; b) R. W. Waller, L. J.
Diorazio, B. A. Taylor, W. B. Motherwell, T. D. Sheppard, Tetrahedron
2010, 66, 6496–6507; c) X. Wang, S.-Y. Wang, S.-J. Ji, Org. Lett. 2013, 15,
1954–1957; d) M. Bachman, S. E. Mann, T. D. Sheppard, Org. Biomol.
Chem. 2012, 10, 162–170; e) S. Marcaccini, R. Pepino, M. C. Pozo, Tetra-
hedron Lett. 2001, 42, 2727–2728; f) F. Leon, D. G. Rivera, L. A. Wessjo-
hann, J. Org. Chem. 2008, 73, 1762–1767.
[
72, 304–307.
[
[
14] N. Kagawa, J. P. Malerich, V. H. Rawal, Org. Lett. 2008, 10, 2381–2384.
15] a) Y. Tamaru, M. Kimura, Pure Appl. Chem. 2008, 80, 979–991; b) B. M.
Trost, J. Quancard, J. Am. Chem. Soc. 2006, 128, 6314–6315.
16] M. Ishikura, H. Kato, Tetrahedron 2002, 58, 9827–9838.
[
[
[
17] D. Das, S. Roy, Adv. Synth. Catal. 2013, 355, 1308–1314.
18] a) S. Cacchi, G. Fabrizi, Chem. Rev. 2005, 105, 2873–2920; b) M. Bandini,
A. Eichholzer, Angew. Chem. 2009, 121, 9786–9824; Angew. Chem. Int.
Ed. 2009, 48, 9608–9644.
[30] V. G. Nenajdenko, A. L. Reznichenko, E. S. Balenkova, Tetrahedron 2007,
63, 3031–3041.
[31] W. P. Unsworth, G. Coulthard, C. Kitsiou, R. J. K. Taylor, J. Org. Chem.
2014, 79, 1368–1376.
[
[
[
[
19] a) T. D. Montgomery, Y. Zhu, N. Kagawa, V. H. Rawal, Org. Lett. 2013, 15,
[32] D.-J. Cheng, S.-K. Tian, Adv. Synth. Catal. 2013, 355, 1715–1718.
[33] B. Alcaide, P. Almendros, R. Rodrꢄguez-Acebes, J. Org. Chem. 2006, 71,
2346–2351.
1
140–1143; b) J. Chen, M. J. Cook, Org. Lett. 2013, 15, 1088–1091.
20] L. Du, P. Cao, J. Xing, Y. Lou, L. Jiang, L. Li, J. Liao, Angew. Chem. 2013,
25, 4301–4305; Angew. Chem. Int. Ed. 2013, 52, 4207–4211.
21] T. Hoshi, K. Sasaki, S. Sato, Y. Ishii, T. Suzuki, H. Hagiwara, Org. Lett. 2011,
3, 932–935.
22] Y. Zhu, V. H. Rawal, J. Am. Chem. Soc. 2012, 134, 111–114.
1
1
Received: June 12, 2014
Published online on August 29, 2014
Chem. Eur. J. 2014, 20, 13375 – 13381
www.chemeurj.org
13381 ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim