Studies of the formation of all-carbon quaternary centres, en route to lyngbyatoxin A. A comparison of phenyl and 7-substituted indole systems.

Article abstract of DOI:10.1039/b401490a

Copper mediated allylic substitutions and conjugate additions to geranyl, cinnamyl and allylic indole compounds have been investigated with the aim of finding a method for the creation of the all-carbon quaternary centre present in the natural product lyngbyatoxin A. Reaction conditions have been found giving a 68% SN2' selectivity in the copper mediated addition of PhMgBr to geranyl chloride, as well as 99% and 95% SN2' selectivity in the copper catalysed addition of EtMgBr to cinnamyl chloride and acetate, respectively. When the optimised reaction conditions were applied to the corresponding allylic compounds containing a 7-substituted indole moiety, the regioselectivity was reversed giving only the SN2 product. The allylic indole-containing substrates were also found to be unproductive in Pd- or Mo-catalysed SN2'-type substitution reactions. In related studies, copper catalysed conjugate addition of EtMgBr to the tricyclic lactam 6-methyl-pyrrolo[3,2,1-ij]quinolin-4-one gave a maximum of 20% of the 1,4-addition product.

Full text of DOI:10.1039/b401490a

View Article Online / Journal Homepage / Table of Contents for this issue
Studies of the formation of all-carbon quaternary centres, en route
to lyngbyatoxin A. A comparison of phenyl and 7-substituted
indole systems
Janne E. Tønder, Masood Hosseini, Alex B. Ahrenst and David Tanner*
Department of Chemistry, Technical University of Denmark, Building 201, Kemitorvet,
DK-2800 Kgs. Lyngby, Denmark. E-mail: dt@kemi.dtu.dk
Received 2nd February 2004, Accepted 11th March 2004
First published as an Advance Article on the web 20th April 2004
Copper mediated allylic substitutions and conjugate additions to geranyl, cinnamyl and allylic indole compounds
have been investigated with the aim of finding a method for the creation of the all-carbon quaternary centre present
in the natural product lyngbyatoxin A. Reaction conditions have been found giving a 68% SN2Ј selectivity in the
copper mediated addition of PhMgBr to geranyl chloride, as well as 99% and 95% SN2Ј selectivity in the copper
catalysed addition of EtMgBr to cinnamyl chloride and acetate, respectively. When the optimised reaction conditions
were applied to the corresponding allylic compounds containing a 7-substituted indole moiety, the regioselectivity
was reversed giving only the SN2 product. The allylic indole-containing substrates were also found to be unproductive
in Pd- or Mo-catalysed SN2Ј-type substitution reactions. In related studies, copper catalysed conjugate addition of
EtMgBr to the tricyclic lactam 6-methyl-pyrrolo[3,2,1-ij]quinolin-4-one gave a maximum of 20% of the 1,4-addition
product.
Table 1 Testing CuClؒ2LiCla
Introduction
In our ongoing efforts towards the stereoselective total
synthesis of lyngbyatoxin A (Fig. 1), we have so far focused
on the formation of the all-carbon quaternary centre.1–3
Entry
CuClؒ2LiCl (equiv.)
Addition time (h)
SN2Ј
SN2
1
2
3
4
0.1
0.2
1.05
1.05
2
2
2
22
40
40
42
78
60
60
58
10.5
a Reaction conditions: THF, 0 ЊC, Addition of geranyl chloride.
Products were identified by 1H NMR.
alkylcopper reagent to an indolyl-α,β-unsaturated ester (4) or
tricyclic unsaturated lactam 5.12 All of these strategies would
rely on the use of external chiral ligands to control absolute
stereochemistry, but we decided to start our investigations in
the racemic series. Furthermore, since relatively few methods
have been developed for creation of all-carbon quaternary
centres in which one of the substituents is an aromatic
ring,5,13–16 we decided to include the phenyl analogues in our
exploratory studies of strategies A and B.
Fig. 1
We have previously succeeded in this by a Lewis acid
mediated rearrangement of an appropriately substituted vinyl
epoxide.4 Having experienced a loss of enantiopurity and the
outcome not being completely suitable for gram scale synthesis,
we have however turned our attention towards allylic substi-
tution (SN2Ј) and conjugate addition reactions.
In our initial overall strategy we wished to prepare the
trisubstituted indole 1, in which the quaternary stereocentre is
set and the indolactam ring could be formed subsequently
(Scheme 1).
Results and discussion
1
Allylic substitution with an arylcopper reagent
Since the most convergent strategy is based on the connection
of an indole and the alkyl side chain, we started our program
with a study of allylic substitution by arylcopper species
generated in situ from Grignard reagents, a reaction which has
been investigated previously by Bäckvall et al.5 (for Ar = phenyl
in Scheme 2).
Strategy A is the most appealing approach for obtaining
1 since it involves a retrosynthetic disconnection between the
indole and a geranyl species 2, and the desired C–C bond form-
ation could thus in principle be achieved via allylic substitution
using an indolylcopper reagent.
We started by repeating the work of Bäckvall et al., using
the CuClؒ2LiCl system, and obtained a 22% yield of the
SN2Ј product 6 (Table 1, entry 1), which corresponds well with
the 18% reported.5 Increasing the amount of copper to 0.2
equivalents led to the expected increase in 6 (40%) (Table 1,
entry 2). However, neither increasing the amount of copper
further to 1.05 equivalents nor increasing the addition time
resulted in a better SN2Ј selectivity (Table 1, entries 3, 4).
As the CuClؒ2LiCl system did not seem promising, we tested
a range of different copper salts using the technique of “reverse
addition”, in which geranyl chloride is added to the reaction
It is however known5–7 that the selectivity for such SN2Ј
attack is poor for simple arylcopper reagents, so an alternative
would be strategy B involving attack of an alkylcopper species
on a substituted indole (3). Strategy B could also involve allylic
substitution catalysed by palladium8 or molybdenum9–11 while
strategies C and D are based on conjugate addition of an
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1447

View Article Online
Scheme 1 Strategies for creation of the quaternary centre.
Table 2 Effect of leaving groupa
Entry
Leaving group
SN2Ј
SN2
1
2
3
Cl
Br
OAc
67
59
0
33
41
100
Scheme 2 Allylic substitution with an aryl copper reagent to give the
SN2Ј product 6 or the SN2 product 7.
a Reaction conditions: 1.05 eq. CuCN, THF, 25 ЊC, 10.5 h addition of
geranyl compound. Products were identified by 1H NMR.
mixture containing the arylcopper magnesium reagent
(Scheme 2). We observed that SN2Ј-selectivity increased in the
following order; CuCl < CuOTf < CuBr < CuI ≤ CuCN and
that preformation of a stoichiometric arylcopper species is
necessary for a high SN2Ј selectivity, since 0.2 equivalents of
CuCN gave rise to only 9% of the SN2Ј product (6). In addition,
if no copper salt was added only the SN2 product (7) was
observed. In the following we have focused on the regio-
selectivity of the reactions, but the isolated yields were
consistently higher than 80%.
Examination of the leaving group on the geranyl species
revealed that Cl > Br > OAc (Table 2), and in accordance with
the observations published by Bäckvall’s group it was found
the addition of geranyl chloride over a shorter period (3.5 h vs.
10.5 h) gave a lower selectivity.
Table 3 Effect of temperaturea
Entry
Temperature (ЊC)
Solvent
SN2Ј
SN2
1
2
3
4
5
6
Ϫ30
20
40
Ϫ50
0
20
THF
THF
THF
CH2Cl2
CH2Cl2
CH2Cl2
12
68
66
67
68
63
88
32
34
33
32
37
a Reaction conditions: 1.05 eq. CuCN, 10.5 h addition of geranyl
chloride. The Grignard reagents were formed in THF. Products were
identified by 1H NMR.
2
Allylic substitution with an alkylcopper reagent
Interesting solvent effects were also observed (Table 3):
dichloromethane and THF were equally good, but the results
obtained in THF were very temperature-dependent, with
an optimum around 20 ЊC, whereas selectivities in dichloro-
methane were consistently high (67–68%) at temperatures
ranging from Ϫ50 to 20 ЊC.
Since the indolylcopper species were not useful nucleophiles in
the allylic substitution, we then opted to incorporate the indole
moiety in the allylic electrophile, and from our previous work4
we knew that Heck reactions in the 7-position of the indole
gave easy access to the required unsaturated system.
By variation of the described reaction conditions an
optimum of 68% SN2Ј selectivity was found using phenyl-
magnesium bromide, 1.05 eq. CuCN, 10.5 h addition time
of geranyl chloride, 20 ЊC, and THF or dichloromethane as
solvent. To our knowledge this is the best SN2Ј selectivity ever
obtained for this substrate in a reaction involving arylcopper
nucleophiles derived from Grignard reagents.17
Unfortunately, subjecting the Grignard reagent formed
from N-allyl18 (8) or N-tosyl4 (9) protected 7-bromoindole
to the optimised conditions gave mostly the reduced indole
resulting from simple protonation of the indolylcopper
or indolylmagnesium bromide reagent, and only traces of the
substitution product could be identified by GC-MS.
Starting from unprotected and N-benzyl protected 7-bromo-
indole, 4a and 4b were prepared by a modified literature
procedure (Scheme 3).4 The trans nature of the double bond
was established by NOE experiments with the allylic alcohols
(10a, 10b and 10c) showing NOE-effects between the methyl
and the methylene group on the double bond.
The tosyl protected compound (4c) was more difficult to
obtain, as the Heck reaction with N-tosylated 7-bromoindole
gave a mixture of isomers, which could neither be separated as
the α,β-unsaturated ester (4c), the allylic alcohol (10c) nor as
the allylic acetate (13c). We, therefore, attempted to prepare 4c
by tosylation of 4a, (NaH, TsCl) but the desired product was
formed in only low yields (<20%), and the major product was
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1448

View Article Online
Scheme 3 Synthesis of indole substrates. a: R = H, b: R = Bn, c: R = Ts. i) NaH, 0
pyridine 0 20 ЊC. iv) Ac2O, pyridine, 20 ЊC. v) ClCO2Me, pyridine, 0 20 ЊC.
20 ЊC. ii) Dibal-H, toluene, Ϫ50
20 ЊC. iii) (EtO)2POCl,
tricycle 5 (Scheme 3). In fact, 5 could be formed in quantitative
yields by treatment of 4a with NaH alone, giving an
overall yield of 59% from 7-bromoindole compared to a
literature report of a 12% yield from indoline.12 Since this
α,β-unsaturated lactam could be an interesting starting point
for conjugate addition, it was included in the studies. By a
selective reduction with Dibal-H the allylic alcohols (10b and
10c) were prepared in good yield. From this point a range of
different leaving groups could be introduced. In order to
prepare the phosphate of the benzyl protected indole, 10b
was treated with diethyl chlorophosphate. However, only the
corresponding chloride (11b), arising by chloride substitution
of the initially formed phosphate group was isolated. Subject-
ing 10c with the electron withdrawing tosyl protecting group to
the same conditions afforded the desired phosphate (12c) along
with 3% of the chloride.
methane at low temperatures (Table 4, entries 3, 4, 7, and 8),
which corresponds well with previous findings.13 THF also gives
selectivities above 90% at low temperatures (Table 4, entries
16 and 17) and this is convenient for cases in which
dichloromethane is incompatible with the Grignard reagent.
The solvent used for formation of the Grignard reagent (THF
or diethyl ether) did not significantly influence the results.
In the study conducted by Alexakis et al.,13 it was found that
cinnamyl acetate favoured the SN2 product under the con-
ditions applied (Ϫ80 ЊC, CH2Cl2, triethyl phosphite). However,
since the indole acetates were easily prepared we wanted to find
reaction conditions favouring the SN2Ј-product when an acetate
was used as a leaving group. To our satisfaction, a high SN2Ј-
selectivity was achieved when the reactions were performed at
higher temperatures (Table 5, entries 1, 5, 7, 10, and 11). Again
the best solvent was dichloromethane, although this time the
solvent for the Grignard reagent was very important, with
diethyl ether providing significantly higher selectivities than
THF. This is in line with a previous study by Bäckvall et al.17
who used substrates which did not contain the cinnamyl moiety.
In order to test whether or not an extended addition time
(5 h) would raise the selectivity, two experiments were performed
under the optimum conditions determined; no significant
change in selectivity was observed (Table 4, entry 9 and Table 5,
entry 6).
The acetates 13bc and the methyl carbonate 14 were prepared
by standard procedures by treatment with acetic anhydride or
methyl chloroformate and pyridine.
Aware of the structural similarity between the quaternary
centre in the natural product sporochnol (synthesised recently
by Hoveyda et al.14) and lyngbyatoxin A, we initiated model
studies and were able to reproduce the excellent regioselectivity
reported in the copper catalysed addition of diethylzinc to
diethyl cinnamyl phosphate.14 To our surprise, however, the
indole system (in 12c and 13bc) once again proved to be
recalcitrant and no reaction occurred either with or without the
addition of CuCN.
We then turned to the copper catalysed reactions with alkyl
Grignards; for the initial investigation and optimisation of
reaction conditions we used the commercially available
cinnamyl chloride and cinnamyl acetate together with ethyl-
magnesium bromide (Scheme 4).
Testing the conditions on the indole system showed yet again
an unexpected lack of reactivity, as only 11b and 12c afforded
products arising from substitution (Table 6, Scheme 5). In both
cases the SN2 product (18) was predominant and only traces of
SN2Ј-product (17) could be detected.
Scheme 4 Allylic substitution with an alkylcopper reagent (Tables 4
and 5).
Scheme 5 Allylic substitution with indole electrophiles.
By variation of temperature and solvent, we found that the
best conditions for copper catalysed allylic substitution using
cinnamyl chloride were to perform the reactions in dichloro-
In order to test if this reversal of selectivity in the indole
systems was caused by increased steric hindrance, we syn-
thesised the corresponding phenyl-analogue β-methylcinnamyl
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1449

View Article Online
Table 4 Reaction between EtMgBr and cinnamyl chloride (Scheme 4)a
Entry
Reaction solvent
Grignard solvent
Temperature (ЊC)
SN2Ј
SN2
1
2
3
4
5
6
7
8
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Et2O
Et2O
Et2O
Et2O
THF
THF
THF
THF
THF
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
THF
THF
THF
THF
20
0
Ϫ30
Ϫ78
20
82
90
94
97
86
92
97
99
98
87
92
68
84
71
87
92
94
18
10
6
3
14
8
3
1
2
13
8
32
16
29
13
8
0
Ϫ30
Ϫ78
Ϫ78
20
9b
10c
11
12
13
14
15
16
17
0
Ϫ30
Ϫ78
20
0
Ϫ30
Ϫ78
THF
THF
THF
6
a Reaction conditions: 10% CuCN, 20% triethyl phosphite, 30 min addition time plus 10 min reaction. Products were identified by GC. b 5 h addition
time. c Without addition of triethyl phosphite the relationship is 80 : 20.
Table 5 Reaction between EtMgBr and cinnamyl acetate (Scheme 4)a
Entry
Reaction solvent
Grignard solvent
Temperature (ЊC)
SN2Ј
SN2
1
2
3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Et2O
Et2O
Et2O
Et2O
THF
THF
THF
THF
THF
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
THF
THF
THF
THF
20
0
Ϫ30
Ϫ78
20
20
0
Ϫ30
Ϫ78
20
89
47
6
11
53
94
95
6
6
5
21
54
13
14
67
52
93
94
96
93
4
5
5
94
94
95
79
46
87
86
33
48
7
6b
7
8
9
10
11
12
13
14
15
16
17
0
Ϫ30
Ϫ78
20
0
Ϫ30
Ϫ78
THF
THF
THF
6
4
7
a Reaction conditions: 10% CuCN, 20% triethyl phosphite, 30 min addition time plus 10 min reaction. Products were identified by GC. b 5 h addition
time.
Table 6 Allylic substitution with indole electrophilesa
Compound
R
X
Reaction solvent
Temperature (ЊC)
SN2Ј
SN2
11
Bn
Bn
Ts
Bn
Bn
Ts
Cl
Cl
Et2O
0
Ϫ78
20
0
20
0
Ϫ78
<5
—
<5
—
—
>95
11
CH2Cl2
CH2Cl2
Et2O
CH2Cl2
CH2Cl2
CH2Cl2
—
12
OPO(EtO)2
OAc
OAc
OAc
OCO2Me
62b
13b
13b
13c
14
—
—
—
—
c
—
d
Bn
20
—
a EtMgBr (in ether), 50% CuCN, 100% triethyl phosphite, 30 min addition time plus 24 h reaction. Products were identified by 1H NMR. b 33% of the
allylic alcohol 10c. c Quantitative reaction with the acetate giving the allylic alcohol 10c as the sole product. d Decomposition of starting material.
acetate (19) by Dibal-H reduction19 of ethyl β-methylcinnamate
followed by a standard acetylation (Scheme 6).
Performing the allylic substitution, employing the optimum
conditions determined for cinnamyl acetate (Table 5, entry 7),
we observed only a slight loss of SN2Ј-selectivity (from 95% to
90%). The complete reversal of selectivity seen in the indole
series is, therefore, not caused solely by the extra methyl group
on the double bond as compared to cinnamyl acetate.
Having synthesised the allylic indole compounds we then
tested the molybdenum catalysed allylic alkylation, which has
proven to be very selective towards the branched product when
cinnamyl methyl carbonate was used as electrophile.9 We were
able to reproduce the excellent selectivity reported for the
cinnamyl system, but for the indole system in 14, no reaction
occurred (Scheme 7).
Scheme 6 Allylic substitution with β-methylcinnamyl acetate.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1450

View Article Online
Table 7 Conjugate addition to α,β-unsaturated esters (Scheme 8)a
Compound
R
23 : 24 : 25
Yield (%)
4ab
4ac
4ad
4b
H
H
H
Bn
Ts
0 : 79 : 21
0 : 0 : 0
0 : 78 : 22
0 : 0 : 0
82
—
82
—
—
4c
0 : 0 : 0
a Reaction conditions see Scheme 8. Products were identified by 1H
NMR. b 2.0 equivalents EtMgBr. c 2.0 equivalents EtMgBr and CuCN.
d 2.0 equivalents EtMgBr and no CuCN.
Table 8 Conjugate addition to 5 (Scheme 10)a
Scheme 7 Molybdenum catalysed allylic alkylation.
Entry
Solvent
Additives
Conversion (%)
The use of Pd(PPh3)4 as a precatalyst was equally unsuccess-
ful, and only the more reactive complex PdCl(allyl)PPh3 20 gave
a trace of the substituted product. Molecular modeling21,22 of
the (η3-allyl)palladium complex (Fig. 2) indicates that it is
very strained and, indeed, unlikely to be formed. In addition,
the modelling also shows that the N-benzyl protecting group
efficiently blocks the approach of any external nucleophile to
the face of the π-allyl system opposite the palladium.
1
2
3
4
5
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
THF
THF
THF
Et2O
Et2O
—
P(OEt)3 b
—
13
19
—
11
—
7
b
P(OEt)3 b, TMSClc
P(OEt)3 b, BF3(OEt)2
P(OEt)3 b, DMPUc
P(OEt)3 b, HMPAc
P(OEt)3
c
6
7d
8e
9
P(OEt)3b, TMSClc
—
f
P(OEt)3 b, TMSClc, HMPAc
P(OEt)3
12
—
—
—
17
7
b
10
11
12
13
14
15
16
P(OEt)3 b, TMSClc
P(OEt)3 bb, TMSClc, HMPAc
P(OEt)3
P(OEt)3 b, TMSClc
P(OEt)3 b, HMPAc
P(OEt)3 b, TMSClc, HMPAc
Et2O
Et2O
1
20
a Reaction conditions: 1 eq. EtMgBr (in ether), 0.5 eq. CuCN, rt, 24 h.
Products were identified by GC. b 1 eq. c 2 eq. d EtMgBr (in THF) was
used. e 1.5 eq. of EtMgBr (in ether). f Decomposition of starting
material.
equivalent of Grignard reagent to deprotonate the indole
nitrogen. Interestingly, we now obtained the 1,2-addition
product 24a but also 21% of a tricyclic product (25) with the
desired quaternary centre. In this case it is possible that the
alkyl group is delivered intramolecularly via a precoordination
of the Grignard reagent to the indole nitrogen. It was not
possible to suppress the formation of 24a by increasing the
amount of CuCN and, surprisingly, an experiment without
CuCN gave the same product distribution as with CuCN
(Table 7).
The tricyclic compound (25) could be formed by two
pathways (Scheme 9). The first possibility is a 1,4-attack on
the α,β-unsaturated ester (E)-4a followed by ring closure
during aqueous work up. The second possibility is a trans/cis
isomerisation after the initial deprotonation of the indole
nitrogen. The cis isomer (Z)-4a is then ready for a ring closure
followed by a 1,4-addition.
Fig. 2 Low energy conformation of the (η3-allyl)palladium complex
with two PPh3 ligands. Calculated using Macromodel 8.0.21
3
Conjugate addition
We then turned our attention to the copper catalysed conjugate
addition of an alkyl Grignard reagent to the α,β-unsaturated
esters 4bc with ethylmagnesium bromide being chosen as a
simple model (Scheme 8 and Table 7).
To our surprise, no reaction with the N-protected indoles
occurred, not even 1,2-addition to give 24bc and at present we
have no convincing explanation for this lack of reactivity. We
then tried to use the unprotected compound 4a with an extra
Encouraged by the formation of 25 we tried the conjugate
addition to the tricycle using the same conditions as before
(Scheme 10), and satisfyingly we again saw the formation of 25
(Table 8, entry 2). In an experiment without copper, no reaction
was observed.
Scheme 8 Conjugate addition to α,β-unsaturated esters (Table 7).
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1451

View Article Online
Scheme 9 Possible mechanism for the formation of 25.
Experimental
Tetrahydrofuran and diethyl ether were distilled under nitrogen
from sodium/benzophenone prior to use. Dichloromethane was
distilled under nitrogen from calcium hydride prior to use.
Methyl methacrylate, ethyl crotonate, and Ti(OiPr)4 were
vacuum distilled before use. All other commercially available
compounds were used as received. Dichlorobis(tri-o-tolyl-
Scheme 10 Conjugate addition to unsaturated lactam 5 (Table 8).
phosphine)palladium(),24
tetrakis(triphenylphosphine)pal-
ladium(0),25 chloro(allyl)(triphenylphosphine)palladium(),20
(S,S)-N,NЈ-bis(2-pyridinecarboxamide)-1,2-cyclohexane,26
N-allyl-7-bromoindole,18 N-tosyl-7-bromoindole,4 4b,4 and
10b4 were synthesised following literature procedures. All
air- and moisture-sensitive reactions were carried out under
argon in oven-dried glassware. For prolonged addition times a
syringe pump was used. It was not possible to obtain good
microanalyses of the indole compounds due to their instability,
Despite the low yield, we felt the result was promising enough
to warrant further investigation, as detailed in Table 8. To
our disappointment, however, only low yields (0–20%) were
obtained, the best conditions being diethyl ether or dichloro-
methane as solvent with addition of triethyl phosphite, TMSCl,
and HMPA. Besides the conditions listed in Table 8, we also
tried other copper sources (CuI, CuBr, Cu(OTf )2) and another
preformed cuprate (Bu2CuLi) in combination with the additives
referred to above. None of these experiments gave rise to any of
the desired product.
1
so these were characterised by HRMS. H (300 or 500 MHz)
and 13C (75 or 125 MHz) NMR spectra were recorded on a
Varian Mercury 300 or a Varian Inova 500, respectively. IR
spectra were recorded on a Perkin Elmer 1600 series FT-IR.
High resolution mass spectra were provided by the Department
of Chemistry, University of Copenhagen.
Conclusion
Although the experiments involving indole moieties (in either
the nucleophilic or the electrophilic components of the reac-
tions studied) were unsuccessful, our investigations involving
the corresponding phenyl compounds have led to several new
findings. Thus, the regioselectivity obtained in the allylic substi-
tution of geranyl chloride with a an arylcopper reagent repre-
sents a significant improvement on earlier results5 and, as far as
we are aware, the 68 : 32 ratio in favour of the SN2Ј product is
the best yet recorded for this substrate. Furthermore, synthetic-
ally useful reaction conditions for SN2Ј substitutions involving
cinnamyl chloride (99% SN2Ј) and cinnamyl acetate (95% SN2Ј)
have been identified. The results involving cinnamyl acetate are
especially valuable, since this substrate has not previously been
reported to favour SN2Ј substitution. The efficient formation of
the tricyclic lactam 5 in two steps from 7-bromoindole is also
noteworthy, and this type of reaction sequence should provide a
rapid entry to tricyclic alkaloid systems e.g. hippadine.23
The large differences in reactivity between the phenyl and
7-substituted indole (for the range N-H, N-Bn or N-Ts)
substrates were unexpected, and our results underline the
limitations of some contemporary methods for the formation
of all-carbon quaternary centres. Since the strategies outlined
above for the formation of the desired quaternary centre in
lyngbyatoxin A were unsuccessful, we are currently exploring
routes based on allylic substitution using appropriately
substituted aryl species which will allow formation of the indole
moiety at a later stage in the synthesis.
General procedure for allylic substitutions with arylcopper
reagents
Bromobenzene (60 µl, 0.57 mmol) was added over a period
of 30 min (with a syringe pump) to a suspension of Mg (28 mg,
1.1 mmol) in THF (1 ml), and the reaction mixture was stirred
vigorously at 20 ЊC for 30 min. The solution of phenyl-
magnesium bromide was then added over a period of 1 h (with
a syringe pump) to a suspension of CuCN (51 mg, 0.57 mmol)
in THF or dichloromethane (2 ml) at 0 ЊC. Then a solution
of geranyl chloride (100 µl, 0.54 mmol) in THF or dichloro-
methane (1 ml) was added over 10.5 h at the chosen tem-
perature. When the addition was complete, the mixture was
stirred for another hour before quenching with a buffer solution
(NH4Cl/NH4OH (sat), pH = 8–9, 3 ml). The mixture was
extracted with dichloromethane (3 × 10 ml), dried (MgSO4),
and evaporated to give ≈110–115 mg of a yellow oil. The
products (6 and 7) could be isolated by column chromatography
(pentane/diethyl ether 95 : 5). Spectral data were in accordance
with those previously published.5
General procedure for allylic substitutions with alkylcopper
reagents
Reaction with cinnamyl chloride. To a suspension of CuCN
(7 mg, 0.08 mmol) in solvent (5 ml) was added triethyl
phosphite (0.03 ml, 0.17 mmol) and the mixture was stirred at
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1452

View Article Online
room temperature for 15 min until the CuCN was fully dis-
solved. The mixture was then cooled to the appropriate temper-
ature and cinnamyl chloride was added (0.10 ml, 0.72 mmol).
To the mixture was added EtMgBr (0.80 mmol) dropwise over
30 min, and the reaction mixture was stirred for an additional
10 min. The products were identified by 1H NMR and the
distribution was analysed by GC. Spectral data were in
accordance with those previously published (1514 and 1627).
18.8, 8.6. IR (KBr, νmax/cmϪ1): 3420, 2968, 2934, 1026, 946, 797,
730. MS: m/z (EI) 243.1620 (Mϩ. C16H21NO requires 243.1623).
1
25: H NMR (CDCl3) δ 7.67 (d, J = 3.6 Hz, 1H), 7.47 (dd,
J = 1.0, 7.8 Hz, 1H), 7.26 (t, J = 7.6 Hz, 1H), 7.17 (dd, J = 0.9,
7.4 Hz, 1H), 6.71 (d, J = 3.6 Hz, 1H), 2.90 (d, J = 16.5 Hz, 1H),
2.81 (d, J = 16.5 Hz, 1H), 1.75 (m, 2H), 1.43 (s, 3H), 0.81 (t,
J = 7.5 Hz, 3H). 13C NMR (CDCl3) δ 158.5, 148.5, 132.0, 127.5,
124.9, 123.4, 123.0, 121.4, 121.1,118.0, 110.1, 17.4. IR (KBr,
νmax/cmϪ1): 2965, 1718, 1440, 1390, 1304, 801, 735. MS: m/z (EI)
213.1161 (Mϩ. C14H15NO requires 213.1154), 213, 184, 154.
Reaction with cinnamyl acetate. To a suspension of CuCN
(5 mg, 0.06 mmol) in solvent (5 ml) was added triethyl
phosphite (0.02 ml, 0.11 mmol) and the mixture was stirred
at room temperature for 15 min until the CuCN was fully
dissolved. The mixture was then cooled to the appropriate
temperature and cinnamyl acetate was added (0.10 ml,
0.60 mmol). To the mixture was added EtMgBr (0.65 mmol)
dropwise over 30 min, and the reaction mixture was stirred for
an additional 10 min. The products were identified by 1H NMR
and the distribution was analysed by GC. Spectral data were in
accordance with those previously published (1514 and 1627).
General procedure for the copper catalysed conjugate addition of
EtMgBr to 5
To a suspension of CuX (50 mol%) in solvent (THF, dichloro-
methane or Et2O) (5 ml) at room temperature was added
the appropriate additives and 5 (100 mol%, 50–100 mg). The
reaction was stirred at room temperature for 15 min and
EtMgBr was added dropwise over 10 min. The mixture was
kept stirring for 24 h, whereafter it was quenched with water
(5 ml), extracted with DCM (3 × 5 mL) and dried (MgSO4).
The solvent was evaporated and the extent of conversion was
determined by 1H NMR on the crude product.
Reaction with allylic indoles 11–14. To a suspension of CuCN
(50 mol%) in diethyl ether or dichloromethane (5 ml) at room
temperature was added diethyl phosphite (100 mol%). The
reaction was allowed to stir for 15 min until all CuCN was
dissolved. The mixture was then cooled to the appropriate
temperature and the indole substrate (100 mol%) and EtMgBr
(100 mol%) were added with an addition time of 30 min. The
mixture was kept stirring for 24 hours and quenched with water
(5 ml) and extracted with dichloromethane (3 × 5 ml), dried
(MgSO4) and evaporated. Product distribution was determined
25: See above.
Indole chemistry
(E )-Ethyl 3-(indol-7-yl)-crotonate (4a). PdCl2(P(o-tol)3)2
(400 mg, 0.51 mmol), 7-bromoindole (2.00 g, 10.3 mmol),
triethylamine (25 ml, 180 mmol) and tetrabutylammonium
bromide (800 mg, 2.48 mmol) were added to a solution of ethyl
crotonate (30 ml, 241 mmol) in DMF (100 ml) and stirred for
120 min at 90 ЊC. The mixture was cooled to room temperature,
diluted with ethyl acetate (100 ml) and washed with brine
(200 ml). The aqueous layer was extracted with diethyl ether
(3 × 100 ml) and the combined organic phases were dried
(MgSO4) and evaporated to a dark liquid, which, in addition to
product, contained traces of DMF. After three purification
steps by column chromatography (ethyl acetate/hexane 1 : 10)
1.42 g (60%) of the desired product was obtained.
1
by crude H NMR on the crude product. The SN2Ј products
could not be isolated due to the small amount formed.
18b 1H NMR (CDCl3) δ 7.55 (d, J = 7.8 Hz, 1H), 7.23
(m, 3H), 7.07 (t, J = 7.3 Hz, 1H), 7.03 (d, J = 3.2 Hz, 1H), 6.88
(dd, J = 1.5, 7.2 Hz, 1H), 6.86 (d, J = 6.1 Hz, 2H), 6.57 (d,
J = 3.2 Hz, 1H), 5.47 (br s, 2H), 5.36 (tq, J = 1.5, 7.2 Hz, 1H),
2.08 (br s, 2H), 1.80 (s, 3H), 1.34 (m, 2H), 0.90 (t, J = 7.3 Hz,
3H). 13C NMR (CDCl3) δ 139.1, 133.7, 132.9, 130.6, 129.8,
128.4, 127.0, 126.1, 126.4, 124.3, 122.9, 120.3, 119.3, 102.3,
51.2, 30.4, 22.4, 19.2, 13.9. MS: m/z (EI) 289.1835 (Mϩ.
C21H23N requires 289.1830), 289, 198, 168, 91.
1H NMR (CDCl3) δ 8.47 (br s, 1H), 7.38 (d, J = 7.7 Hz, 1H),
7.25 (t, J = 2.7 Hz, 1H), 7.19 (dd, J = 1.1, 7.3 Hz, 1H), 7.15 (t,
J = 7.6 Hz, 1H), 6.60 (dd, J = 2.1, 3.2 Hz, 1H), 6.27 (q, J = 1.3
Hz, 1H), 4.26 (q, J = 7.2 Hz, 2H), 2.69 (s, 3H), 1.35 (t, J = 7.2
Hz, 2H). 13C NMR (CDCl3) δ 167.1, 155.2, 132.5, 128.7, 126.8,
124.8, 121.4, 120.6, 119.7, 118.1, 102.9, 60.1, 19.7, 14.3. IR
(KBr, νmax/cmϪ1): 3448, 1654, 1420. MS: m/z (EI) 229.1109
(Mϩ. C14H15NO2 requires 229.1103), 229, 184, 154, 77.
1
18c H NMR (CDCl3) δ 7.64 (d, J = 3.8 Hz, 1H), 7.42 (d,
J = 8.4 Hz, 2H), 7.35 (dd, J = 1.4, 7.7 Hz, 1H), 7.15 (t, J = 7.5
Hz, 1H), 7.09 (d, J = 7.9 Hz, 2H), 7.03 (dd, J = 1.4, 7.4 Hz, 1H),
6.67 (d, J = 3.7 Hz, 1H), 5.21 (tq, J = 1.3, 7.1 Hz, 1H), 2.31
(s, 3H), 2.12 (q, J = 7.4 Hz, 2H), 2.02 (d, J = 1.3 Hz, 3H), 1.48
(m, 2H). 13C NMR (CDCl3) δ 148.1, 144.3, 137.0, 135.4, 135.2,
133.4, 130.8, 129.4, 129.1, 127.5, 126.9, 124.3, 119.9, 111.2,
30.6, 22.8, 21.7, 18.7, 14.3. MS: m/z (EI) 353.1454 (Mϩ.
C21H23NO2S requires 353.1450).
(E )-Ethyl 3-(N-p-toluenesulfonylindol-7-yl)-crotonate (4c). p-
Toluenesulfonyl chloride (3.0 g, 15.7 mmol) was added to a 0 ЊC
solution of 4a (906 mg, 3.95 mmol) in THF (50 ml), and the
mixture was stirred for 5 min. Then NaH (55% in mineral oil,
200 mg, 4.0 mmol) was added, and the greyish mixture was
stirred for 30 h at 0 ЊC. Column chromatography (ethyl acetate/
heptane 1 : 9) afforded 214 mg (14%) of the product as a
red/brown oil. Recovered starting material (760 mg, 84%).
1H NMR (CDCl3) δ 7.64 (d, J = 3.8 Hz, 1H), 7.45 (dd, J = 1.3,
7.8 Hz, 1H), 7.42 (d, J = 8.4 Hz, 2H), 7.20 (t, J = 7.6 Hz, 1H),
7.13 (d, J = 8.4 Hz, 1H), 7.06 (dd, J = 1.3, 7.5 Hz, 1H), 6.71 (d,
J = 3.8 Hz, 1H), 5.64 (q, J = 1.3 Hz, 1H), 4.22 (q, J = 7.1 Hz,
2H), 2.58 (d, J = 1.3 Hz, 3H), 2.32 (s, 3H). 13C NMR (CDCl3)
δ 166.7, 158.4, 144.7, 134.9, 133.6, 132.9, 132.1, 130.8, 129.6,
126.9, 126.1, 124.4, 121.5, 118.7, 111.4, 60.0, 21.8, 21.5, 14.6. IR
(KBr, νmax/cmϪ1): 3159, 2976, 1713, 1638, 1348, 1189, 1172,
1128, 1092, 801, 720, 680. MS: m/z (EI) 383.1173 (Mϩ.
C21H21NO4S requires 383.1191), 383, 184, 154, 91.
General procedure for the copper catalysed conjugate addition of
EtMgBr to 4abc
To a suspension of CuCN (0–100 mol%) in dichloromethane
(5 ml) at room temperature was added triethyl phosphite
(0–100 mol%) and the mixture was allowed to stir for 15 min.
until CuCN was fully dissolved. To this mixture was added
the indole substrate 4abc (100 mol%, ≈100 mg) and EtMgBr
(100–200 mol%) dropwise over 10 min. The mixture was
allowed to stir at room temperature for 24 hours and quenched
with water (5 ml) and extracted with dichloromethane
(3 × 5 ml). 1H NMR provided the product distribution.
1
24a: H NMR (CDCl3) δ 8.33 (bs, 1H), 7.54 (d, J = 7.4 Hz,
1H), 7.23 (t, J = 2.7 Hz, 1H), 7.11 (d, J = 7.3 Hz, 1H), 7.07 (dd,
J = 1.5, 7.3 Hz, 1H), 6.58 (dd, J = 2.3, 3.2 Hz, 1H), 5.70 (d,
J = 1.2 Hz, 1H), 2.36 (d, J = 1.2 Hz, 3H), 1.74 (q, J = 7.5 Hz,
4H), 1.02 (t, J = 7.5 Hz, 6H). 13C NMR (CDCl3) δ 136.3, 134.1,
130.0, 128.3, 124.3, 124.7, 120.4, 120.0, 119.5, 103.2, 54.7, 34.4,
6-Methyl-pyrrolo[3,2,1-ij]quinolin-4-one (5). NaH (55% in
mineral oil, 300 mg, 12.5 mmol) was added to a solution of 4a
in THF (200 ml) at 0 ЊC. The ice bath was removed, and stirring
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1453

View Article Online
was continued for 2 h. Then the reaction was quenched with
water (200 ml) and evaporated. Water (50 ml) and diethyl ether
(50 ml) were added, and extraction with diethyl ether (5 × 50
ml), drying (MgSO4) of the combined organic phases and
evaporation gave 234 mg (98%) of the desired product as a light
green solid.
111.1, 64.2, 64.1, 63.7, 63.6, 21.4, 18.9, 16.1, 16.0. IR (KBr,
νmax/cmϪ1): 3448, 2984, 1371, 1271, 1176, 1032, 802, 682. MS:
m/z (EI) 477.1359 (Mϩ. C23H28NO6PS requires 477.1375), 477,
258, 168, 99.
In addition the corresponding chloride (E)-3-(N-p-toluene-
sulfonylindol-7-yl)-but-2-en-1-chloride was isolated (6 mg,
0.02 mmol, 3%).
1H NMR (CDCl3) δ 8.42 (d, J = 3.7, 1H), 7.92 (d, J = 3.5,
1H), 7.83 (d, J = 7.6, 1H), 7.42 (t, J = 7.6, 1H), 6.67 (d, J = 7.6,
1H), 6.53 (d, J = 1.2, 1H), 2.57 (d, J = 1.2, 3H). 13C NMR
(CDCl3) δ 159.0, 149.0, 132.5, 127.5, 124.9, 123.4, 122.9, 121.4,
121.1, 110.1, 17.4. IR (KBr, νmax/cmϪ1): 2923, 1664, 1626, 1599,
1448, 1316, 1118, 808. MS: m/z (EI) 183.0676 (Mϩ. C12H9NO
requires 183.0684), 183, 154, 127, 77. Mp: 142–143 ЊC.
1H NMR (CDCl3) δ 7.63 (d, J = 3.7 Hz, 1H), 7.41 (d, J = 8.2
Hz, 2H), 7.40 (dd, J = 7.8, 1.3 Hz 1H), 7.18 (t, J = 7.5 Hz, 1H),
7.11 (d, J = 8.0 Hz, 2H), 7.04 (dd, J = 7.4, 1.3 Hz, 1H), 6.69 (d,
J = 3.7 Hz, 1H), 5.48 (tq, J = 7.7, 1.4 Hz, 1H), 4.26 (d, J = 7.5
Hz, 2H), 2.30 (s, 3H), 2.15 (d, J = 1.2 Hz, 3H).
(E )-1-Acetyloxy-3-(N-benzylindol-7-yl)-but-2-ene
(13b).
(E )-3-(N-p-Toluenesulfonylindol-7-yl)-)-but-2-en-1-ol (10c).
Dibal-H (1.2 M in toluene, 6.0 ml, 7.2 mmol) was added
dropwise to a Ϫ78 ЊC solution of 4c in toluene (60 ml), and
the reaction mixture was stirred for 3 h during which the
temperature gradually reached 20 ЊC. The mixture was diluted
with dichloromethane (100 ml) and Rochelle’s salt (sat. 100 ml)
was added. The aqueous layer was extracted with dichloro-
methane (3 × 100 ml) and the combined organic phases were
dried (MgSO4) and evaporated. Column chromatography (ethyl
acetate/hexane 1 : 1) afforded 444 mg (65%) of the desired
product as colourless crystals.
Pyridine (0.090 ml, 1.1 mmol) and acetic anhydride (0.10 ml,
1.0 mmol) were added to a solution of 10b (203 mg, 0.73 mmol)
in dichloromethane (5 ml) and stirred for 24 hours at room
temperature. The reaction mixture was quenched with H2O
(5 ml) and dichloromethane (5 ml), extracted with dichloro-
methane (3 × 10 ml) and the combined organic phases were
dried (MgSO4) and evaporated. Column chromatography (ethyl
acetate/heptane 2 : 3) afforded 210 mg (90%) of the desired
product as a pale brown oil.
1H NMR (CDCl3) δ 7.63 (dd, J = 1.2, 7.9 Hz, 1H), 7.32–7.22
(m, 3H), 7.15–7.07 (m, 2H), 6.86 (dd, J = 1.2, 7.2 Hz, 1H), 6.77
(m, 2H), 6.65 (d, J = 3.2 Hz, 1H), 5.55 (dt, J = 1.5, 6.9 Hz, 1H),
5.48 (br s, 2H), 4.72 (br s, 2H), 2.01 (s, 3H), 1.85 (s, 3H). 13C
NMR (CDCl3) δ 205.5, 170.3, 139.8, 139.5, 131.1, 130.8, 129.3,
128.6, 127.1, 126.2, 124.8, 122.1, 120.0, 119.4, 102.1, 60.8, 51.4,
20.2, 19.0. IR (KBr, νmax/cmϪ1): 3031, 1739, 1420, 1361, 1313,
1231, 1028, 796, 725. MS: m/z (EI) 319.1573 (Mϩ. C21H21NO2
requires 319.1572), 319, 244, 168, 154, 91.
1H NMR (CDCl3) δ 7.60 (d, J = 3.5, 1H), 7.44 (d, J = 8.0,
2H), 7.38 (d, J = 7.7, 1H), 7.18 (t, J = 6.8, 1H), 7.13 (d, J = 8.0,
2H), 7.07 (d, J = 7.4, 1H), 6.68 (d, J = 3.8, 1H), 5.63 (t, J = 8.1,
1H), 4.33 (br s, 1H), 2.31 (s, 3H), 2.12 (s, 3H). 13C NMR
(CDCl3) δ 144.4, 140.6, 134.5, 133.8, 133.3, 132.7, 130.7, 129.3,
126.8, 126.6, 126.5, 124.3, 120.3, 111.6, 59.2, 21.4, 18.4. IR
(KBr, νmax/cmϪ1): 3422, 1369, 1169, 1090, 802, 740, 687. MS:
m/z (EI) 341.1095 (Mϩ. C19H19NO3S requires 341.1086), 341,
186, 168, 154, 91. Mp: 132–133 ЊC.
(E )-1-Acetyloxy-3-(N-p-toluenesulfonylindol-7-yl)-but-2-ene
(13c). Pyridine (0.10 ml, 1.2 mmol) and acetic anhydride
(0.10 ml, 1.1 mmol) were added to a solution of 10c (233 mg,
0.68 mmol) in dichloromethane (10 ml) and stirred for 7 hours
at room temperature. The reaction was quenched with water
(10 ml) and extracted with dichloromethane (3 × 10 ml) and the
combined organic phases were dried (MgSO4) and evaporated
to give 244 mg (93%) of the desired product as a light yellow oil,
which on standing turned into a low melting point solid.
1H NMR (CDCl3) δ 7.65 (d, J = 3.8 Hz, 1H), 7.42 (d, J = 8.4
Hz, 2H), 7.39 (dd, J = 1.3, 7.7 Hz, 1H), 7.17 (t, J = 7.5 Hz, 1H),
7.11 (t, J = 8.1 Hz, 2H), 7.05 (dd, J = 1.3, 7.5 Hz, 1H), 6.70 (d,
J = 3.7 Hz, 1H), 5.43 (tq, J = 6.8, 1.4 Hz, 1H), 4.76 (d, J = 6.9
Hz, 2H), 2.31 (s, 3H), 2.12 (s, 3H), 2.11 (s, 3H). 13C NMR
(CDCl3) δ 171.0, 144.5, 142.5, 135.0, 133.2, 132.5, 130.6, 129.2,
126.7, 126.7, 124.1, 123.0, 122.1, 120.4, 111.1, 61.4, 21.4, 21.0,
18.9. IR (KBr, νmax/cmϪ1): 3448, 1736, 1375, 1233, 1176, 802,
681. MS: m/z (EI) 383.1189 (Mϩ. C21H21NO4S requires
383.1191), 383, 258, 244, 168, 154, 91.
(E )-3-(N-Benzylindol-7-yl)-1-chloro-but-2-ene (11b). Diethyl
chlorophosphate (0.060 ml, 0.40 mmol) was added dropwise to
a 0 ЊC solution of 10b (100 mg, 0.31 mmol) and pyridine
(0.10 ml, 1.2 mmol) in dichloromethane (5 ml), the reaction was
stirred at room temperature for 2 hours. Then the mixture was
diluted with diethyl ether (10 ml) and H2O (10 ml), extracted
with diethyl ether (3 × 10 ml) and dried (MgSO4). Column
chromatography (ethyl acetate/heptane 1 : 1) afforded 71 mg
(77%) of the desired product as a pale brown oil.
1H NMR (CDCl3) δ 7.60 (dd, J = 6.7, 1.2 Hz, 1H), 7.25 (m,
3H), 7.10 (m, 2H), 6.86 (m, 3H), 6.63 (d, J = 3.2 Hz, 1H), 5.63
(dt, J = 1.5, 6.7 Hz, 1H), 5.39 (m, 2H), 4.19 (m, 2H), 1.79 (m,
3H). 13C NMR (CDCl3) δ 140.8, 139.0, 132.7, 130.8, 130.5,
128.8, 127.3, 126.3, 125.6, 122.3, 121.9, 120.3, 119.5, 102.4,
51.9, 40.8, 19.3. IR (KBr, νmax/cmϪ1): 3426, 3030, 2922, 1420,
1314, 1176, 796, 725. MS: m/z (EI) 295.1130 (Mϩ. C19H18ClN
requires 295.1128), 295, 259, 168, 91.
(E )-3-(N-p-Toluenesulfonylindol-7-yl)-but-2-en-1-yl diethyl
phosphate (12c). Pyridine (0.10 ml, 1.2 mmol) and diethyl
chlorophosphate (0.19 ml, 1.3 mmol) were added to a 0 ЊC
solution of 10c in dichloromethane (10 ml). The mixture was
heated to room temperature and stirred for 20 h and quenched
with water (10 ml), extracted with dichloromethane (3 × 10 ml)
and the combined organic phases were dried (MgSO4) and
evaporated. Column chromatography (ethyl acetate/hexane
1 : 1) afforded 260 mg (93%) of the desired product as a
colourless oil.
(E )-3-(N-Benzylindol-7-yl)-1-methoxycarbonyloxy-but-2-ene
(14). Methyl chloroformate (0.20 ml, 2.6 mmol) was added
dropwise to a 0 ЊC solution of 10b (380 mg, 1.37 mmol)
in pyridine (2 ml) and dichloromethane (2 ml). The ice bath
was removed, and stirring was continued at room temperature
for 15 h. Dichloromethane (10 ml) and water (10 ml) were
added, and extraction with dichloromethane (3 × 10 ml), drying
of the combined organic phases (MgSO4), and evaporation
followed by column chromatography (ethyl acetate/hexane 1 : 8)
afforded 341 mg (74%) of the desired product as a colourless
oil.
1H NMR (CDCl3) δ 7.64 (d, J = 3.7 Hz, 1H), 7.41 (d, J = 8.4
Hz, 2H), 7.39 (dd, J = 7.6, 1.4 Hz, 1H), 7.17. (t, J = 7.5 Hz, 1H),
7.10 (d, J = 8.0 Hz, 2H), 7.04 (dd, J = 7.4, 1.3 Hz, 1H), 6.69 (d,
J = 3.8 Hz, 1H), 5.46 (dq, J = 1.4, 6.7 Hz, 1H), 4.75 (t, J = 7.1
Hz, 2H), 4.17 (q, J = 7.1 Hz, 4H), 2.30 (s, 3H), 2.11 (s, 3H), 1.36
(t, J = 7.1 Hz, 6H). 13C NMR (CDCl3) δ 144.0, 142.5, 134.7,
133.1, 132.9, 130.6, 129.3, 126.6, 124.1, 122.6, 122.4, 120.5,
1H NMR (CDCl3) δ 7.65 (dd, J = 1.6, 7.9 Hz, 1H), 7.28 (m,
3H), 6.92 (dd, J = 1.5, 7.4 Hz, 1H), 6.89 (d, J = 5.8 Hz, 2H),
6.66 (d, J = 3.2 Hz, 1H), 5.59 (tq, J = 1.4, 7.0 Hz, 1H), 5.49 (d,
J = 12.7, 2H), 4.79 (d, J = 5.0 Hz, 2H), 3.80 (s, 3H), 1.87 (s, 3H).
13C NMR (CDCl3) δ 156.1, 141.1, 139.1, 132.4, 130.7, 130.4,
129.0, 128.8, 127.4, 126.3, 123.7, 122.5, 120.3, 119.6, 102.5,
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1454

View Article Online
64.7, 55.0, 51.8, 19.8. IR (KBr, νmax/cmϪ1): 2955, 1747, 1444,
1314, 1266. 941, 795, 725. MS: m/z (EI) 335.1514 (Mϩ.
C21H21NO3 requires 335.1521), 335, 244, 168, 91.
5 J. E. Bäckvall, E. S. M. Persson and A. Bombrun, J. Org. Chem.,
1994, 59, 4126.
6 T. Fujisawa, S. Iida, H. Yukizaki and T. Sato, Tetrahedron Lett.,
1983, 24, 5745.
7 C. C. Tseng, S. D. Paisley and H. L. Goering, J. Org. Chem., 1986,
51, 2884.
ꢀ-Methylcinnamyl acetate (19)
8 T. Cuvigny, M. Julia and C. Rolando, J. Organomet. Chem., 1985,
285, 395.
9 B. M. Trost and I. Hachiya, J. Am. Chem. Soc., 1998, 120, 1104.
10 A. V. Malkov, P. Spoor, V. Vinader and P. Kocovský, Tetrahedron
Lett., 2001, 42, 509.
11 A. V. Malkov, I. R. Baxendale, D. J. Mansfield and P. Kocovský,
J. Chem. Soc., Perkin Trans. 1, 2001, 1234.
12 U. Franke and E. Röder, Arch. Pharm., 1976, 309, 185.
13 A. Alexakis, C. Malan, L. Lea, C. Benhaim and X. Fournioux,
Synlett, 2001, S1, 927.
14 C. A. Luchaco-Cullis, H. Mizutani, K. E. Murphy and
A. H. Hoveyda, Angew. Chem., Int. Ed., 2001, 40, 1456.
15 J.-Y. Legros and J.-C. Fiaud, Tetrahedron Lett., 1990, 31, 7453.
16 D. Rodríguez, J. P. Sestelo and L. A. Sarandeses, J. Org. Chem.,
2003, 68, 2518.
Pyridine (1.0 ml, 12 mmol) and acetic anhydride (1.0 ml,
11 mmol) were added to a solution of β-methyl cinnamyl
alcohol19 (300 mg, 2.3 mmol) in dichloromethane (10 ml) and
stirred for 12 hours at room temperature. The reaction was
quenched with water (10 ml) and extracted with dichloro-
methane (3 × 10 ml) and the combined organic phases were
dried (MgSO4) and evaporated. Column chromatography (ethyl
acetate/hexane 1 : 7) afforded the product as colourless oil.
(370 mg, 1.9 mmol, 96%). Spectral data were in accordance
with those previously published.28
20 and 21
17 J. E. Bäckvall, M. Sellén and B. Grant, J. Am. Chem. Soc., 1990, 112,
6615.
The synthesis was performed according to the general procedure
for allylic substitutions with alkylcopper reagents (cinnamyl
acetate) using β-methylcinnamyl acetate in dichloromethane at
20 ЊC.
18 A. P. Dobbs, K. Jones and K. T. Veal, Tetrahedron Lett., 1997, 38,
5379.
19 R. Martin, G. Islas, A. Moyano, M. A. Pericas and A. Riera,
Tetrahedron, 2001, 57, 6367.
20 J. Powell and B. L. Shaw, J. Chem. Soc. A, 1967, 1839.
21 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp,
M. Lipton, C. Caufield, G. Chang, T. Hendrickson and W. C. Still,
J. Comput. Chem., 1990, 11, 440.
22 H. Hagelin, B. Åkermark and P.-O. Norrby, Organometallics, 1999,
18, 2884.
23 D. L. Boger and S. E. Wolkenberg, J. Org. Chem., 2000, 65, 9120 and
references therein.
24 T. E. Nielsen and D. Tanner, J. Org. Chem., 2002, 67, 6366.
25 D. R. Coulson, Inorg. Synth., 1972, 13, 121.
26 D. A. Conlon and N. Yasuda, Adv. Synth. Catal., 2001, 343, 137.
27 G. W. Kabalka, D. Tejedor, N.-S. Li, R. R. Malladi and S. Trotman,
Tetrahedron, 1998, 54, 15525.
28 T. Morimoto, M. Hirano and T. Koyama, J. Chem. Soc., Perkin
Trans. 2, 1985, 1109.
29 I. Fleming, S. K. Patel and C. J. Urch, J. Chem. Soc., Perkin Trans. 1,
1989, 115.
30 L. A. Paquette and G. D. Maynard, J. Org. Chem., 1989, 54, 5054.
The spectral data were in accordance with those previously
published (2029 and 2130)
Acknowledgements
We thank the Danish Research Agency, DTU, and Novo
Nordisk A/S for financial support, Prof. Per-Ola Norrby for
the molecular mechanics calculations, and Dr Jørgen Øgaard
Madsen for obtaining mass spectra.
References
1 K. Fuji, Chem. Rev., 1993, 93, 2037.
2 E. J. Corey and A. Guzman-Perez, Angew. Chem., Int. Ed., 1998, 37,
388.
3 J. Christoffers and A. Mann, Angew. Chem., Int. Ed., 2001, 40, 4591.
4 J. E. Tønder and D. Tanner, Tetrahedron, 2003, 59, 11571.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 4 4 7 – 1 4 5 5
1455