Lewis acid catalyzed Nenitzescu indole synthesis

Article abstract of DOI:10.1002/jhet.5570430410

A novel method for Lewis acid catalyzed Nenitzescu indole syntheses of 5-hydroxyindoles bearing different substituents in positions 1 (Alk, Bn, Ar), 2 (Me, Et, Ph), and 3 (COOEt, COMe, CONHPh) as well as tricyclic derivatives are reported. The method is simple, rapid, efficient, and allows preparation of hydroxyindoles from 1,4-benzoquinone and enamines in good to excellent yields with the use of low-polar solvents in the presence of weak Lewis acids catalysts. The formation of 5-hydroxyindoles under such mild conditions is explained in terms of a non-redox mechanism.

Full text of DOI:10.1002/jhet.5570430410

Jul-Aug 2006
Lewis Acid Catalyzed Nenitzescu Indole Synthesis
873
Valeriya S. Velezheva*, Albert G. Kornienko, Sergey V. Topilin, Ascar D. Turashev,
Alexander S. Peregudov, and Patrick J. Brennana
A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 28 Vavilov Str,
119991 GSP-1 Moscow, Russia
Fax: (7-095)135-5085.
E-mail: vel@ineos.ac.ru
a Department of Microbiology, Immunology and Pathology, Colorado State University, Fort Collins, Colorado
80523-1682 U.S.A
Fax: (970) 491-1815.
E-mail: Patrick.Brennan@ColoState.edu
Received August 5, 2005
3
3
COR
R2
HO
COR
R2
O
ZnI2
+
CH2Cl2
HN
R1
N
R1
O
3 a-n
2 a-n
A novel method for Lewis acid catalyzed Nenitzescu indole syntheses of 5-hydroxyindoles bearing
different substituents in positions 1 (Alk, Bn, Ar), 2 (Me, Et, Ph), and 3 (COOEt, COMe, CONHPh) as
well as tricyclic derivatives are reported. The method is simple, rapid, efficient, and allows preparation of
hydroxyindoles from 1,4-benzoquinone and enamines in good to excellent yields with the use of low-polar
solvents in the presence of weak Lewis acids catalysts. The formation of 5-hydroxyindoles under such mild
conditions is explained in terms of a non-redox mechanism.
J. Heterocyclic Chem., 43, 873 (2006).
Introduction.
simplest synthetic entry into 5-hydroxyindole-based key
intermediates of pharmacologically active molecules and
drugs [11,12], including those cited above. For example,
the facile synthesis of an advanced E 09 intermediate
comprises five steps including the Nenitzescu indol-
ization, in contrast to the 15-step synthesis starting from
3-chlorophenol [13]. The reaction, as well as Fisher
indole synthesis, plays a prominent role in classical indole
ring formation due to readily available precursors. The
Fisher indolization, with rare exceptions, proceeds
smoothly and furnishes good yields of target products but
is very seldom suitable for preparation of polysubstituted
5-hydroxyindoles [14]. On the contrary, the Nenitzescu
reaction, in principle, provides the possibility of solving
the latter problem to a great extent. However the reaction
has been fickle since at times it proceeds with good yields
of indoles while at other times yields are low or the
reaction may even fail altogether [15]. It was noted that
the Nenitzescu reaction is highly affected by the nature of
the substituents on the starting compounds and the
reaction medium [5,7,9,15,16]. The 5-hydroxy derivatives
of 3-acylbenzofurans are often obtained instead of the
corresponding 3-acyl-5-hydroxyindoles. In many cases,
the simultaneous formation of indole and benzofuran
derivatives has been observed [15]. The scope of the
reaction has mainly been restricted to the synthesis of 5-
hydroxyindoles with 3-carboxylic ester groups. With
5-Hydroxyindoles are found in nature and display a
wide variety of biological activities [1,2]. A series of
polysubstituted 5-hydroxyindole-based agents have been
developed recently including the antiviral and immuno-
modulatory drug arbidol, ethyl 6-bromo-4-dimethyl-
aminomethyl-5-hydroxy-1-methyl-2-phenylthio-methyl-
indole-3-carboxylate hydrochloride [3] (and its 4-subst-
ituted derivatives [4]), as well as compounds of potential
value as drugs. For example, N-benzyl substituted
analogues of the anti-inflammatory drug indometacin and
indometacin-related N-benzyl indoleacetamide exhibit
selective cyclooxygenase (COX-2) inhibitory [5] and
multidrug resistance related-protein
1
(MRP-1)
modulatory activities [6,7], respectively. The antitumor
agent E 09 (a fully synthetic indolequinone with 1,2,3,5-
substituent pattern) and its analogues are good substrates
for human DT-diaphorase which possibly is a molecular
target for enzyme-directed bioreductive drug development
[8,9]. Syntheses of medicinally interesting 2,4-diamino-
8H-pyrimido[4,5-b]indol-6-ols were carried out recently
via the Nenitzescu reaction [10].
The discovery of the above synthetically challenging
and medicinally useful compounds has intensified the
search for new effective drugs with broad clinical
applications. The Nenitzescu reaction followed by
functional group interconversions has proved to be the

874
V. S. Velezheva, A. G. Kornienko, S. V. Topilin,
A. D. Turashev, A. S. Peregudov, and P. J. Brennan.
Vol 43
those, however, being at other times, low [5,7,9,10,16-
18]. The corresponding N-benzylaminoacrylic esters
when reacted with benzoquinone in nitromethane gave 2-
methyl-1-benzyl-5-hydroxyindole-3-carboxylates in 47%
yield and its 2-ethyl homologue in 53% yield [7].
Moreover, under such conditions the reaction goes slowly
(16 - 48 hr) and fails altogether with enaminoanilides as
enamino components in CH2Cl2 [19]. Even as lately as
2002, ethyl 5-hydroxy-2-propyl-indole-3-carboxylate was
reported to be obtained by the Nenitzescu reaction in
acetic acid, in 6% yield only [12]. Thus, large-scale
production of the Nenitzescu indoles should be expensive.
Previously, the application of the Nenitzescu reaction was
somewhat limited due to not only low yields of 5-
hydroxyindoles, but also difficulties in their isolation as
Since the Nenitzescu reaction includes two steps of
nucleophilic addition (conjugation of an enamino component
to quinone and ring closure due to the nitrogen-to-carbonyl
carbon addition) one could expect that, as in other similar
nucleophilic addition reactions, the application of acidic
catalysis would facilitate both of them. We believed that
Lewis acids would not impede the interaction of enamines
with benzoquinones. We also expected Lewis acids to be
most effective for processes conducted in low-polar solvents.
At present, nitromethane and AcOH are considered as the
best solvents for the Nenitzescu reaction. In these solvents,
however, the process occurs via an oxidation-reduction
pathway generally marked by the formation, along with 5-
hydroxyindoles, of many by-products.
Results and Discussion.
pure compounds because of
contamination with
We chose the interaction of 3-methylaminocrotonate 2a
with quinone 1 furnishing 5-hydroxyindole 3a as the first
prototypical reaction. We did obtain high purity indole 3a
in good yield (> 80%) from the reaction carried out in
benzene in the presence of catalytic amounts of ZnCl2
(Table 1, Scheme 1).
numerous by-products including those of characteristic
dark red colour [15].
Recently, advances in the efficiency of the Nenitzescu
indole synthesis were made. These came from the choice of
proper solvents and the application of a solid-phase method
described by Ketcha et al. [16]. The authors were the first
to solve the problem of regioselective synthesis of 1,2,3,6-
pentasubstituted 5-hydroxyindoles via the reaction of
mono-substituted quinones with the corresponding
enaminoamides. However, this solid-phase method
employing nitromethane as the solvent and applied for the
preparation of only 5-hydroxyindole 3-carboxamides, does
not seem entirely satisfactory since it is a multi-stage
procedure, employs hazardous and expensive chemicals
and, in some cases, gives low yields of target 5-
hydroxyindoles. Hence the problem of 5-hydroxyindole
yields still remains topical. All of this underlines the need
for a flexible strategy that could reduce the reaction time,
provide good yields of 5-hydroxyindoles and use
inexpensive reagents and solvents.
Scheme 1
3
3
HO
COR
R2
COR
R2
O
ZnI2
+
CH2Cl2
N
R1
HN
R1
O
1
2 a-n
3 a-n
2,3a: R1=R2=Me, R3=OEt
b: R1=H, R2=Me, R3=OEt
h: R1+R2=(CH2)5, R3=OEt
i: R1=H, R2=Me, R3=Me
c: R1=CH2Ph, R2=Me, R3=OEt j: R1= R2= R3=Me
d: R1=Ph, R2=Me, R3=OEt
e: R1=Me, R2=Et, R3=OEt
f: R1=H, R2=Ph, R3=OEt
g: R1=Me, R2=Ph, R3=OEt
k: R1 = CH2Ph, R2= R3=Me
l: R1=Ph, R2= R3=Me
m: R1=4-Br-Ph, R2= R3= Me
n: R1= CH2Ph, R2=Me, R3=NHPh
In order to develop improved methods we dwelt on
catalytic interactions of 1,4-benzoquinone 1 with
enaminoesters, enaminoketones and enaminoanilide
2a-n, typical enamino components for the reaction. As
to the Lewis acid catalyzed reactions of enaminoesters
with 1,4-benzoquinones, only those of N-substituted
aminofumarates have been reported by Domschke et al.
[20]. These aminofumarates react with 1,4-benzo-
To assess how catalysts and their quantities, as well as
how solvents affect the reaction we also used other Lewis
acids (AlCl3, BF3·Et2O, and ZnI2) as well as a protic acid,
CF3CO2H (TFA), and solvents (toluene, carbon
tetrachloride, methylene dichloride). Our experiments
demonstrated that in the absence of catalyst, the yields of
indole 3a in benzene and CH2Cl2 achieved 30 and 21%,
respectively, with the isolated compound purity being <
.
quinone in diethyl ester under BF3 Et20 catalysis to
afford N-substituted 5-hydroxyindole-2,3-dicarboxyl-
ates in 21 – 90% yields. Engler et al. extensively used
Lewis acid promotion under addition of enol esters and
styrenes to benzoquinones and their imines in
syntheses of either 2-aryl-2,3-dihydrobenzofurans or 2-
aryl-2,3-dihydroindoles, pterocarpanes, and highly
substituted tetrahydro-ꢀ- or ꢁ-carbolines, and also,
benzofuran analogs [21].
1
98% as shown by H NMR. Earlier, Kucklender et al.
obtained a mixture of indole 3a with the corresponding
hydroquinone adduct on heating a benzene solution of the
same species [22]. In all cases when we used our method
for producing indole 3a (and later on indoles 3b-n) a
benzene or CH2Cl2 solution of enamine was gradually
added to a boiling solution of an equimolar amount of

Jul-Aug 2006
Lewis Acid Catalyzed Nenitzescu Indole Synthesis
875
contained in the slightly colored mother liquor obtained
after separation of the product in the form of a white
powder. If the reaction time was limited to 2 hours the yield
was reduced to 80%.
quinone 1 in the same solvent after introduction of a
catalyst. The reaction mixture was then boiled for
additional 40 - 60 minutes (method A). We obtained pure
indole 3a in good yields (80 – 87%) from the reaction
carried out in the presence of catalytic amounts of ZnCl2
or ZnI2 (1 – 10 mol %) (Table 1, entries 3-6, 15-16).
The application of Lewis acid catalysis allowed
preparation of indoles under conditions milder than those
used before. With that, there were neither 5,5-
dihydroxydiindoles, nor hydroquinone, etc., characteristic
of the classical Nenitzescu reaction performed in polar
solvents and believed to proceed via the oxidation-
reduction pattern [15]. Without the catalysts, the reaction
either did not go at the low temperatures or produced
indole 3a in small yield at heating. When passing from
CH2Cl2 to both CCl4 and toluene, with ZnI2 in amount of 8
mol %, the yield decreased to 19 and 67%, respectively.
We believe this to have occurred since the reaction
mixture was more heterogeneous in these solvents and
underwent resinification to a greater extent. When the
reaction was carried out by procedure A with AlCl3 or
BF3·Et2O as catalysts, the yields of indole 3a turned out to
be moderate. The samples of the crude indole 3a listed
under entries 1, 2, 10-14 and 17-19 in Table 1 contained
some impurities and their melting points were below that
of the pure compounds mentioned under entries 3-9, 15,
16 and 20-22.
The ZnI2 catalyzed reaction effected by procedure A
was also employed for the synthesis of indoles 3b-n
bearing different substituents at 1, 2, and 3 positions of
the indole ring (Scheme 1, Table 2). With all enamines
2a-n, the process went smoothly and the direct product
crystallization occurred as the reaction progressed. In all
cases the best yields of indoles 3a-n (between 62 and
95%) resulted when equimolar amounts of the quinone
and enamine were taken.
Table 1
Yields of indole 3a depending on catalysts and reaction conditions.
Entry
Solvent
t[a],
min
Mp, (°C)
Catalyst MXn
(mol %)
Yield
%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
C6H6
CH2Cl2
C6H6
CH2Cl2
C6H6
C6H6
C6H6
C6H6
C6H6
C6H6
45 200-203
40 197-199
None
None
30
21[b]
81
40
60
40
40
40
40
45
60
208-209
208-209
208-209
208-209
208-209
208-209
207-208
196-198
ZnCl2 (8)
ZnCl2 (8)
ZnCl2 (1)
ZnCl2 (5)
ZnCl2 (10)
ZnCl2 (20)
ZnCl2 (50)
ZnCl2 (200)
ZnCl2 (400)
ZnCl2 (0.1)
ZnCl2 (0.2)
ZnI2 (0.2)
ZnI2 (10)
ZnI2 (1)
ZnCl2 (8)
ZnCl2 (8)
AlCl3 (1)
AlCl3 (8)
BF3.Et2O (1)
BF3.Et2O (8)
87
82
81
80
80
78
64[b]
37[b]
42[b]
55[b]
68[b]
83
C6H6
C6H6
105 195-198
40 195-198
300 195-198
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CCl4
toluene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
40
40
40
197-199
207-208
207-208
83
40 180-190
45 195-198
120 177-178
19[b]
67[b]
65[b]
65
60
60
40
206-207
207-208
205-206
45
66
[a] method A. [b] the crude indole 3a.
We found the yield of indole 3a to be appreciably
dependent on the amount of the catalyst applied. In benzene,
high yields of the product, 80-82%, were achieved when
ZnCl2 was used in amounts of 1-20% (Table 1). An increase
in the amount to 50 mol % did not virtually influence the
yield (78%) whereas it was reduced to 64% with 200 mol %
of the catalyst. With 4-fold excess, the yield was noticeably
less - 37%. The reaction responded almost similarly to a
decrease in the amount of ZnCl2 to 0.1 mol % producing the
product in 42% yield. With this catalyst, the highest indole
3a yield of 87% was reached when quinone 1 and enamine
2a were heated in CH2Cl2 in the presence of 8 mol % for 1
hour. We observed a drop in the yield to 55% with a
decrease in amount of the catalyst to 0.2 mol %. The same
proved to be valid for the ZnI2/CH2Cl2 system; reducing the
amount of the catalyst from 8 to 0.2 mol % lowered the
yield from 83 to 68%. With that, prolongation of the heating
of the components from 40 minutes up to several hours did
not noticeably affect the reaction. However, the best yield
(92%) of 3a was gained when the process was conducted in
CH2Cl2 in the presence of ZnI2 at -45° - -30° C for 15 hours
(method B). Additional amount of the indole 3a was
Table 2
ZnI2 catalyzed Nenitzescu Indole 3a-n Synthesis.
Product
Yield %
[a]
Mp (°C)
[c]
3a
83
6027
208-209
]
92[b
3b
3ꢀ
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
77
95
69
67
72
63
78
79
87
77
78
76
62
5327
4427
5327
204-205
197-198
203-204
172-174
174-175
205-207
195-197
294-295
247-248
243-244
230-231
292-293
220-221
4628
5330
3030
5329
[a] method A; [b] method B; [c] max. reported.

876
V. S. Velezheva, A. G. Kornienko, S. V. Topilin,
A. D. Turashev, A. S. Peregudov, and P. J. Brennan.
Vol 43
All indoles obtained are feasible for further use without
ZnI2 or TFA for 4 h afforded benzofuran 5 in 9% and 70%
yields, correspondingly.
additional purification. In the case of enaminoester 2d, the
yield of 1-phenylindole 3d did not exceed 69% with the
lack, however, of isomeric 6-hydroxyindole characteristic
of the Nenitzescu reaction in AcOH [15,23]. The
reactions with 2i-m gave 3-acetyl-5-hydroxyindoles 3j-n
instead of 5-hydroxybenzofurans typical for the
interaction between benzoquinones and enaminoketones
in acidic media, e.g., AcOH [15,24]. With TFA taken as
an example, we further studied whether protic acid
catalysts could be employed in the process. For the
reaction of enamine 2a with quinone 1, the results turned
out to be temperature dependent. At -50° C to -45° C in
CH2Cl2, only product 4 was formed in 94% isolated yield
whereas a mixture of indole 3a, benzofuran 5, and adduct
4 was obtained on heating (Scheme 2).
Thus the experiments showed TFA to be unsuitable as a
catalyst for the indole synthesis. However the low
temperature reaction in the presence of TFA provided the
possibility to obtain pure adduct 4 and to study its
behavior under conditions of the Lewis acid catalyzed
indolization. This turned out to be important in
elucidating the reaction mechanism.
An internal oxidation-reduction mechanism has been
proposed and is now commonly accepted for the
Nenitzescu reaction [15,25]. It involves the oxidation of
an intermediate Michael adduct of type 4 by a quinone.
Hence, the high yields of indole 3a secured by us in the
ZnCl2 or ZnI2 catalyzed reactions did not fit in with the
redox pattern since adduct 4 either does not react at all or
reacts to a small extent under the catalytic conditions. All
these findings count in favor of the hypothesis that the
catalytic Nenitzescu reaction occurs via a non-redox
mechanism presented in Scheme 3.
Scheme 2
COOC2H5
NHCH3
HO
-50 ÷ -45 oC
CH3
Scheme 3
OH
4
-
-
ZnCl2
O
ZnCl2
+ O
H
H
CF3COOH
C6H6, 80 oC
COOC2H5
+
2a
+
CF3COOH
CH2Cl2
O
H3C
NHCH3
O
1+2a
6
H
H
COOC2H5
HO
HO
COOC2H5
CH3
+
N
+
O
CH3
CH3
HO
COOC2H5
HN
CH3
40 oC
O
H
3a
+
4
+
-
-
O
CH3
ZnCl2
ZnCl2
5
7
8
H
These results prompted us to examine whether indole
3a could be formed from adduct 4 in both the TFA and
Lewis acid promoted reactions and if so, whether it
depended on the presence or absence of benzoquinone
and the catalysts at different temperatures. According to
Allen’s data [15], a hydroquinone adduct was transformed
into the corresponding indole in 55% yield in AcOH in
the presence of 10 mol % benzoquinone via a redox
mechanism. We found that without quinone 1, adduct 4
did not transform into indole 3a in the presence of ZnI2 or
TFA in both cases, on heating and cooling. Moreover
adduct 4 was left intact in the presence of either 10 or
even 100 mol % of quinone 1 at -50°C to -30°C with or
without ZnI2. However, it furnished indole 3a in low
yields (21 - 23%), when heated in benzene with 10 or 100
mol % of quinone 1 in the presence or absence of ZnI2.
The reaction mixture also contained a rather large amount
of brick-red colored products typical for the classical
conditions. At the same time, boiling of a benzene
solution of adduct 4 without quinone 1 in the presence of
HO
COOC2H5
CH3
3a
- H2O, - ZnCl2
N
+
CH3
O
H
-
ZnCl2
The conjugate Michael addition is a reversible process,
so that complexing carbonyl oxygen of substrate 1 by the
metal atom of a Lewis acid can shift the equilibrium
towards intermediate 6 or increase the rate of the direct
reaction. This increases the amount of intermediate 6, which
undergoes fast prototropic isomerization at its imino site to give
intermediate 7 with the enamino moiety. The rearrangement
seems plausible to proceed owing to the absence of any
additional base capable of proton abstraction from the
hydroxydienone site of 6. By contrast, as shown by Bernatek
[26], in the presence of ZnC12 and AcOH, the
corresponding hydroxydienone intermediate (formed from
quinone 1 and acetoacetic ester) is converted to a type 4 Michael
adduct followed by cyclocondensation of the latter into

Jul-Aug 2006
Lewis Acid Catalyzed Nenitzescu Indole Synthesis
877
benzofuran 5, obviously via proton abstraction by AcO- as a
base. The benzofuran synthesis is considered to follow a
non-redox mechanism. Besides, the closing of the 5-
membered ring into intermediate 8 is favored according to
Baldwin’s rules (five-Exo-Tet process) and thus, should
be quick. The finding already mentioned that the yields of
indole 3a were higher in the presence of more weak
Lewis acids (ZnC12 or ZnI2) than those in the presence of
AlCl3 or BF3·Et2O can be attributed to the ability of the
latter to deactivate the enamino component (Table 1). The
same effect of deactivation probably occurred when we
increased the amounts of ZnC12 or ZnI2 up to 4-fold
excess (Table 1).
2e, ethyl-3-aminocinnamate [33] 2f, ethyl-3-methyl-
aminocinnamate [34] 2g, ethyl 2-(hexahydro-2H-azepin-
2-yliden)-acetate [35] 2h, 4-amino-pent-3-en-2-one [36]
2i, 4-methylamino-pent-3-en-2-one [37] 2j, 4-benzyl-
amino-pent-3-en-2-one [38] 2k, 4-anilino-pent-3-en-2-one
[39] 2l, 4-(4-bromophenylamino)-3-penten-2-one [38]
2m, 3-benzylamino-crotonanilide [40] 2n were prepared
according to literature procedures.
General Methods.
1
The H NMR spectra were recorded on a Bruker AMX
– 400 spectrometer at 400 MHz. The chemical shifts are
reported in the ꢀ scale (ppm) basing on the residual proton
signal from (CD3)2SO (ꢀ 2.50).
General Procedure for the Synthesis of 5-Hydroxyindoles
3a-n (Method A).
A general characterization of the target compounds
3a-n was accomplished using H NMR spectroscopy. A
1
salient feature in the spectra of these compounds is the
appearance of characteristic groups of signals: two
doublets (ꢀ ca. 7.3 and 7.4 ppm), a doublet of doublets (ꢀ
ca. 6.7 - 6.8 ppm) or a multiplet (ꢀ ca. 6.4 - 7.8 ppm) that
are assigned to the protons - 4, 7, and 6 of the 5-
hydroxyindole ring and protons of one or two aromatic
rings attached to positions 1 and/or 2 of the ring,
respectively.
To a solution of quinone 1 (1.08 g, 10.00 mmoles) in
CH2Cl2 (30 ml) was added ZnI2 (0.32 g, 1.00 mmole). The
temperature was raised to boiling, and a solution of
enamine 2a-n (10 mmoles ) in CH2Cl2 (20 ml) was added
drop by drop at stirring for 5-10 minutes. The mixture was
then stirred under boiling for 40 minutes and cooled to 0-
5° for 2-3 hours. The precipitated crystals were filtered
off and washed with CH2Cl2 (2 x 1 ml) and acetone (2 x 1
ml) (Scheme 1, Tables 1, 2).
Conclusion.
Ethyl 1,2-dimethyl-5-hydroxy-indole-3-carboxylate (3a).
Thus, one can suggest that, depending on the condi-
tions, the Nenitzescu reaction goes via at least two
different mechanisms. The first of those is valid for the
reactions performed in polar solvents. The alternative is
realized when the reactions are conducted in low-polar
solvents in the presence of weak Lewis acids. Thus we
have defined experimental conditions for the realization
of the Nenitzescu reaction via a second, non-redox mech-
anism that was first theoretically proposed by G. R. Allen
in 1966 [27]. On the basis of the latter, we offer a simple and
rapid method to afford polysubstitited 5- hydroxyindoles
in good to excellent yields based on a catalytic version of
the Nenitzescu reaction and applicable to a variety of
enamines; the method does not require expensive
solvents. The products obtained are of high purity so that
they can be further used without purification.We think
that the method will be useful for those chemists who are
in search of biologically active 5-hydroxyindole derived
compounds.
Method A. To a solution of 1,4-benzoquinone (1.08 g,
10.00 mmoles) in CH2Cl2 (30 ml) was added ZnCl2 (1.11 g,
8.00 mmoles). The resulting mixture was heated to boiling
and then a solution of enamine 2a (1.43 g, 10.00 mmoles)
in CH2Cl2 (20 ml) was added drop by drop at stirring for 5-
10 minutes. The mixture was stirred under boiling for
additional 40 minutes and cooled to 0-5° for 2-3 hours. The
precipitated crystals were filtered off and washed with
CH2Cl2 (2 x 1 ml) and acetone (2 x 1 ml) to afford 2.03 g
(87%) of 3a, mp 208-209° (Table 1, entry 4); 209° according to
[41]; 1H nmr (DMSO-d6): ꢀ 1.33 (t, 3H, CH3), 2.66 (s, 3H,CH3-2),
3.64 (s, 3H, NCH3), 4.24 (q, 2H, CH2CH3), 6.65 (dd, 1H, ArH-6),
7.26 (d, 1H, ArH-7), 7.35 (d, 1H, ArH-4), 8.92 (s, 1H, OH).
Method B. To a solution of 1,4-benzoquinone (1.08 g, 10.00
mmoles) in CH2Cl2 (30 ml) was added ZnI2 (0.32 g, 1.00 mmole).
After cooling the mixture to -30 - -45° a solution of 2a (2.86 g,
20.00 mmoles) in CH2Cl2 (20 ml) was added drop by drop at
stirring for 6 minutes. The mixture was then stirred at -30° for 15
hours. The precipitated crystals were filtered off and washed with
CH2Cl2 (2 x 1 ml) and acetone (2 x 1 ml) to afford 2.14 g (92%) of
the product, mp 208-209°; the 1H nmr spectrum was suitable.
Ethyl 2-(2,6-dioxyphenyl)-3-methylaminocrotonate (4).
EXPERIMENTAL
Materials.
To a solution of 1,4-benzoquinone (2.16 g, 20.00 mmoles) in
CH2Cl2 (15 ml) was added TFA (0.2 ml) and then, after cooling
to -45 - -50°, a solution of methylaminocrotonate 2a (3 ml,
20.00 mmoles) in CH2Cl2 (5 ml) was added drop by drop at
stirring for 5-10 minutes. The mixture was further stirred at -30°
for 17 hours. The precipitated crystals were collected by
filtration and washed with CH2Cl2 (2 x1 ml) and acetone (2 x 1
ml). The product was recrystallized from MeOH to afford 4.70 g
1,4-Benzoquinone, ZnCl2, ZnI2, AlCl3, and BF3·Et2O
were purchased commercially and used as received (ZnCl2
was dried before used). Ethyl 3-methylaminocrotonate
[28] 2a, ethyl 3-aminocrotonate [29] 2b, ethyl 3-
benzylaminocrotonate [30] 2c, ethyl 3-phenylamino-
crotonate [31] 2d, ethyl 3-methylamino-2-pentenoate [32]
1
(94%) of 4, mp 138-139°; 137-138° according to [42]; H nmr

878
V. S. Velezheva, A. G. Kornienko, S. V. Topilin,
A. D. Turashev, A. S. Peregudov, and P. J. Brennan.
Vol 43
(DMSO-d6): ꢀ 1.06 (t, 3H, CH3), 1.66 (s, 3H, =CCH3), 2.89 (d,
3H, NCH3), 3.90 (q, 2H, CH2CH3), 6.31 (d, 1H, ArH-3), 6.43
(dd, 1H, ArH-5), 6.55 (d, 1H, ArH-6), 7.99 (s, 1H, OH), 8.47 (s,
1H, OH), 9.29 (br s, 1H, NH).
6.69 (dd, 1H, ArH-6), 7.29 (d, 1H, ArH-7), 7.37 (d, 1H, ArH-4),
8.95 (s, 1H, OH).
Anal. Calcd. for C14H17NO3: C, 67.94; H, 6.87; N, 5.66.
Found: C, 67.97; H, 7.01; N, 5.64.
Ethyl 5-hydroxy-1-methyl-2-phenyl-indole-3-carboxylate (3g).
Interaction of 1,4-Benzoquinone 1 and Ethyl 3-methylaminocro-
tonate 2a in the presence of TFA on heating.
1
Isolated as a colorless solid, mp 205-207°; H nmr (DMSO-
d6): ꢀ 1.05 (t, 3H, CH2CH3), 3.48 (s, 3H, NCH3), 4.02 (q, 2H,
CH2CH3), 6.78 (dd, 1H, ArH-6), 7.36-7.49 (m, 7H, ArH), 9.10
(s, 1H, OH).
Anal. Calcd. for C18H17NO3: C, 73.22; H, 5.76; N, 4,75.
Found: C 73.00; H 5.68; N 4.74.
To a solution of 1 (2.16 g, 20.00 mmoles) in CH2Cl2 (30 ml)
was added TFA (0.15 ml). The solution was heated to boiling,
and a solution of 2a (2.86 g, 20.00 mmoles) in CH2Cl2 (20 ml)
was added drop by drop under stirring for 6 minutes. The
mixture was then stirred at 40° for 40 minutes and cooled to 0-5°
for 2 hours. The precipitated crystals were collected by filtration
and washed with CH2Cl2 (2 x 1 ml) and acetone (2 x 1 ml). The
solid (2.14 g) contained 27% of indole 3a and 18% of adduct 4
and 52% of benzofuran 5 as shown by the comparison between
the integral intensities of 2-CH3 group signals for 3a (ꢀ 2.66), 5 ꢀ
2.68), and adduct 4 (ꢀ 1.66) in the 1H nmr spectrum. The mother
liquor was dried in vacuo, the solid residue washed with CH2Cl2
(2 x 0.5 ml) and acetone (2 x 0.5 ml) to give ethyl 5-hydroxy-2-
methyl-benzofuran-3-carboxylate 5, isolated yield 0.64 g (14%),
mp 142-144°. 143-144° according to [43]; 1H nmr (DMSO-d6): ꢀ
1.35 (t, 3H, CH3), 2.68 (s, 3H, CH3), 4.31 (q, 2H, CH2CH3), 6.73
(dd, 1H, ArH-6), 7.25 (d, 1H, ArH-7), 7.36 (d, 1H, ArH-4), 9.35
(s, 1H, OH). The filtrate contained unidentified products.
Ethyl 8-hydroxy-2,3,4,5-tetrahydro-1H-azepino[1,2-a]indole-
11-carboxylate (3h).
1
Isolated as a colorless solid, mp 195-197°; H nmr (DMSO-
d6): ꢀ 1.33 (t, 3H, COOCH2CH3), 1.80-4.00 (m, 8H, 4 CH2), 4.24
(m, 4H, 2 CH2), 6.65 (dd, 1H, ArH-6), 7.32 (d, 1H, ArH-7), 7.36
(d, 1H, ArH-4), 8.93 (s, 1H, OH).
Anal. Calcd. for C16H19NO3: C, 70.33; H, 6.96; N, 5.13.
Found: C, 70.25; H, 6.99; N, 5.16.
3-Acetyl-1-benzyl-5-hydroxy-2-methylindole (3k).
1
Isolated as a colorless solid, mp 243-244°; H nmr (DMSO-
d6): ꢀ 2.49 (s, 3H, COCH3), 2.66 (s, 3H, CH3), 5.45 (s, 2H,
CH2Ph), 6.59-7.45 (m, 8H, ArH), 9.03 (s, 1H, OH).
Anal. Calcd. for C18H17NO2: C, 77.42; H, 6.09; N, 5.02.
Found: C, 77.39; H, 6.07; N, 4.99.
Interactions of Adduct 4 with ZnI2 or TFA.
To a solution of 0.50 g (2 mmoles) of aminocrotonate 4 in 80
ml of benzene was added ZnI2 (0.064 g, 0.2 mmoles), and the
reaction mixture was then stirred for 40 minutes at boiling. The
solvent was evaporated in vacuo. The mixture contained 81% of
starting adduct 4 and 9% of benzofuran 5. It was displayed by
the comparative analysis of signals from the 2-CH3 and N-CH3
groups of adduct 4 at ꢀ 1.66 and 2.89 accordingly and that from
the 2-CH3 group of benzofuran 5 at ꢀ 2.68 in the 1H nmr
spectrum. The reaction carried out with the same amounts of 4
and the solvent in the presence of TFA (0.15 ml, 0.2 mmoles)
produced 0.31 g (70% isolated yield) of benzofuran 5 after 4
hours boiling.
3-Acetyl-1-(4-bromophenyl]-5-hydroxy-2-methylindole (3m).
1
Isolated as a colorless solid, mp 243-244°; H nmr (DMSO-
d6): ꢀ 2.49 (s, 3H, COCH3), 2.66 (s, 3H, CH3), 5.45 (s, 2H,
CH2Ph), 6.59-7.45 (m, 8H, ArH), 9.03 (s, 1H, OH).
Anal. Calcd. for C18H17NO2: C, 77.42; H, 6.09; N, 5.02.
Found: C, 77.39; H, 6.07; N, 4.99.
1-Benzyl-5-Hydroxy-2-methylindole-3-carboxanilide (3n).
1
Isolated as a colorless solid, mp 292-293°; H nmr (DMSO-
d6): ꢀ 2.50 (s, 3H, COCH3), 2.57 (s, 3H, CH3), 6.40-7.84 (m, 7H,
ArH), 9.12 (s, 1H, OH).
Anal. Calcd. for C17H14BrNO2: C, 59.30; H, 4.07; N, 4.07.
Found: C, 59.36; H, 4.04; N, 3.99.
Interaction of Adduct 4 with 1,4-Benzoquinone.
To a solution of 0.50 g (2 mmoles) adduct 4 in 80 ml of
benzene was added ZnI2 (0.064 g, 0.2 mmoles) and then, under
boiling and stirring, a solution of 1,4 benzoquinone (0.216 g, 2
mmoles) in benzene (5 ml). The mixture was stirred for
additional 40 minutes, allowed to cool down to room
temperature and left overnight. The precipitated crystals were
collected by filtration and washed with CH2Cl2 (2 x 1 ml) and
acetone (2 x 1 ml) to afford 0.11g (23%) of the crude indole 3a,
mp 198-201°. Under the same conditions in the presence of
0.022 g, (0.20 mmoles) of 1,4 benzoquinone indole 3a was
obtained in 21% yield. Without ZnI2, in both cases, mixtures of
compounds containing indole 3a were obtained as determined
Acknowledgements.
The authors would like to thank Dr. Yu. Lyakhovetsky
for his assistance in speculations concerning the non-
redox Nenitzescu reaction mechanism. Financial
support provided by the U.S. Civilian Research &
Development Foundation (RB 2 2032) and Russian
Foundation for Basic Research (01-03-332518) is
greatfully acknowledged.
1
REFERENCES
by H nmr analysis basing on signals of the 2-CH3 group at ꢀ
2.66 and the N- CH3 group at ꢀ 3.64.
[1] H. Y. Meltzer, Neuropsychopharmacology, 21, 106 (1999);
B. E. Leonard, CNS Drugs, 4, 1 (1995); T. C. R.Vijayalaxmi, C. R.
Thomas, J. R. Reiter and S. H. Terence, J. Clin. Oncol., 20, 2575 (2002);
S. Kaneko, K Okumura , Y. Numaguchi , H. Matsui, K. Murase, S.
Mokuno, I. Morishima, K. Hira, Y. Toki, T. Ito and T. Hayakawa. Life
Sciences, 67, 101 (2000).
Ethyl 2-ethyl 5-hydroxy-1-methyl-indole-3-carboxylate (3e).
1
Isolated as a colorless solid, mp 172-174°; H nmr (DMSO-
d6): ꢀ 1.18 (t, 3H, CH2CH3), 1.35 (t, 3H, COOCH2CH3), 3.13 (q,
2H, CH2CH3) 3.68 (s, 3H, NCH3), 4.25 (q, 2H, COOCH2CH3),

Jul-Aug 2006
Lewis Acid Catalyzed Nenitzescu Indole Synthesis
879
[2] R. D. E. Sewell, in “Introduction to the Principles of Drug
Design and Action”, H. J. Smith ed.; Harwood Academic Publishers,
Amsterdam, B.V., 1998, pp 391-410.
[3] T. A. Gus’kova, Pharmaceutical Chemistry Journal, 35, 527
(2001), Springer Verlag New York; J. A. McCullers, Expert Opin. on
Investigational Drugs, 14, 305 (2005), Ashley Publications; R. G.
Glushkov, T. A. Gu’skova, Pharm. Chem. J. (Engl. Trans.), 33, 115
(1999).
[19] E. K. Panisheva, L. M. Alekseeva, A. S. Shashkov, V. G.
Granik, Chem. Heterocycl. Comp. (Russ.), 8, 1164 (2003).
[20] G. Domschke, J. Prakt. Chem., 311, 807 (1969).
[21] T. A. Engler and J. Wanner, Tetrahedron Lett., 38, 6135
(1997); T. A. Engler, K. O. Lynch, W. Chai and S. P. Meduna,
Tetrahedron Lett., 36, 2713 (1995); T. A. Engler, M. A. Letavic, R.
Iyengar, K. O. LaTessa and J. P. Reddy, J. Org. Chem., 64, 2391 (1999).
[22] U. Kucklander, Tetrahedron., 29, 921 (1973).
[23] V. I. Shvedov, E. K. Panisheva, T. F. Vlasova and A. N.
Grinev, J. Gen. Chem. USSR (Engl. Trans.), 9, 1225 (1973).
[24] G. S. Gadaginamath, R. R. Kavali and S. R. Pujar, Synthetic
Comm., 33, 2285 (2003).
[25] J. J. Li, in Name Reactions: A Collection of Detailed
Reaction Mechanisms, Springer, Berlin, 2002, p. 255.
[26] E. Bernatek and T. Ledaal, Acta Chem. Scand., 12, 2053
(1958).
[27] J. R. Allen, Jr. C. Pidacks, and M. J. Weiss, J. Am. Chem.
Soc., 88, 2536 (1966).
[4] D. Wang, Y. U. De Sheng, F. Qin, L. Fang and P. Gong,
Chinese Chem. Lett., 15, 19 (2004).
[5] Y. Leblanc, W. C. Black, C.-C. Chan, S. Charleson, D.
Delorme, D. Denis, C. Bayly, J. Y. Gauthier, R. Grimm, R. Gordon, D.
Guay, P. Hamel, S. Kargman, C. K. Lau, J. Mancini, M. Ouellet, D.
Percival, P. Roy, K. Skorey, P. Tagari, P. Vickers, E. Wong, L. Xu and
P. Prasit, Bioorg. Med. Chem. Lett., 6, 731 (1996).
[6] A. Roller, O. K. Bahr, J. Streffer, S. Winter, M. Heneka, M.
Deininger, R. Meyermann, U. Naumann, E. Gulbins and M. Weller,
Biochem. and Biophys. Res. Comm., 259, 600 (1999).
[7] A. R. Maguire, S. J. Plunkett, S. Papot, M. Clynes, R. O’
Connor and S. Touhey, Bioorg. Med. Chem., 9, 745 (2001).
[8] R. M. Philips, M. A. Naylor, M. Jaffar, S. W. Doughty, S. A.
Everett, A. G. Breen, G. A. Choudry and I. J. Stratford, J. Med. Chem.,
42, 4071 (1999).
[28] E. Knoevenagel, E. Reinecke, Chem. Ber., 32, 423 (1899).
[29] M. C. Bagley; J. W. Dale, J. W. Bower, Synlett, 1149 (2001).
[30] R. Moehlau, Chem. Ber., 27, 3377 (1894).
[31] S. Coffey, J. K. Thomson, F. J. Wilson, J. Chem. Soc., 856
(1936).
[9] J. M. Pawlak, V. V. Khau, D. R. Hutchinson and M. J.
Martinelli, J. Org. Chem., 61, 9055 (1996).
[10] B. Dotzauer and R. Troschutz, Synlett, 1039 (2004).
[11] H. D. Beall, S. Winski, E. Swann, A. R. Hudnott, A. S.
Cotteril, N. O’Sullivan, S. J. Green, R. Bien, D. Siegel, D. Ross and C. J.
Moody, J. Med. Chem., 41, 4755 (1998).
[12] T. M. Bohme, C. E. Augelli-Szafran, H. Hallak, T. Pugsley,
K. Serpa and R. D. Schwarz, J. Med. Chem., 45, 3094 (2002).
[13] M. A. Naylor, M. Jaffar, J. Nolan, M. A. Stephens, S. Butler,
K. B. Patel, S. A. Everett, G. E. Adams and I. J. Stratford, J. Med.
Chem., 40, 2335 (1997).
[14] R. J. Sundberg, in Indoles. Best synthetic methods series.
Academic Press, New York, pp 54-63 (1996); N. N. Suvorov, V. G.
Avramenko, V. N. Shkilkova and L. I. Zamyshlyaeva, British Patent
1.174.034 (1968); Chem. Abstr., 66814m (1970).
[32] J. Decombe, Ann. Chim., 18, 81 (1932).
[33] M. C. Bagley, C. Brace, J. W. Dale, M. Ohnesorge, N. G.;
Phillips, X. Xiong, J. Bower, J. Chem. Soc. Perkin Trans. 1, 14, 1663
(2002).
[34] C. Goldschmidt, Chem. Ber., 29, 105 (1896).
[35] J-P. Celerier, E. L. Deloisy., G. Hommet., P. Maitte, J. Org.
Chem., 44, 3089 (1979).
[36] B. Singh, G. Y.Lesher, J. Heterocyclic Chem., 27, 2085 (1990).
[37] A. P. Terent'ev, E. A.Viktorova, B. M. Eselson, A. N. Kost,
V. V. Ershov, J. Gen. Chem. USSR (Engl. Transl.), 30, 2402 (1960).
[38] G. R. Cook, L. G. Beholz, J. R. Stille, J. Org. Chem., 59,
3575 (1994).
[39] A. D. Garnovskii, V. L. Abramenko, J. Gen. Chem. USSR
(Engl. Transl.), 55, 1630 (1985).
[40] A. N. Grinev, V. N. Ermakova, I. A Mel’nikova, A. P.
Terent’ev, J. Gen. Chem. USSR (Engl. Transl.), 31, 2146 (1961).
[41] A. N. Kost, L. G. Iudin, E. J Zinchenko, Chem. Heterocycl.
Compd. (Engl. Transl.), 9, 306 (1973).
[15] G. R. Allen, in Org. React., 1973, 20, pp 337-454.
[16] D. M. Ketcha, L. J. Wilson and D. E. Portlock, Tetrahedron
Lett., 41, 6253 (2000).
[17]M. Kinugawa, H. Arai, H. Nishikawa, A. Sakaguchi, T. Ogasa, S.
Tomioka, M. Kasai, J. Chem. Soc., Perkin Trans. 1, 2677 (1995).
[18] J. B. Patrick and E. K. Saunders, Tetrahedron Lett., 20, 4009
(1979).
[42] A. N. Grinev, V. N. Ermakova, A. P. Terent’ev, J. Gen.
Chem. USSR (Engl. Transl.), 32, 1928 (1962).
[43] H. L. Mc. Pherson, B. V. Ponder, J. Heterocyclic Chem., 15,
43 (1978).