A Concise Synthesis of Sterically Hindered 3-Amino-2-Oxindoles

Article abstract of DOI:10.1055/s-2003-42074

A new method for the synthesis of 3-alkyl-3-amino-2-oxindoles is reported. These compounds are prepared in a 3-step procedure using a base-mediated nucleophilic addition of benzylidene-imine protected α-aminoesters to 2-nitrofluorobenzene as the key step. The process provides a variety of 3-alkyl-3-amino-2-oxindole analogs in yields of 1-24%. Yields are highest with alanine, phenylalanine and 2-pyridylalanine as the amino acid starting materials, while 3-pyridylalanine and O-methyltyrosine are less efficiently arylated. Sterically hindered amino acids such as valine and phenylglycine are for all practical purposes, not substrates for the key nucleophilic substitution reaction. The resulting 3-alkyl-3-amino-2-oxindoles are important intermediates for the preparation of drug-like substances. The conversion of alanine ethyl ester to 3-amino-3-methyl-2-oxindole is described.

Full text of DOI:10.1055/s-2003-42074

LETTER
2135
A Concise Synthesis of Sterically Hindered 3-Amino-2-Oxindoles
C
oncise Synth
t
esis of
e
Sterically H
p
indered 3-A
h
mino-2-Oxin
e
doles n J. O’Connor,* Zheng Liu
Department of Chemistry, Bayer Research Center, 400 Morgan Lane, West Haven, Connecticut 06437, USA
Fax +1(203)8122650; E-mail: stephen.oconnor.b@bayer.com
Received 13 June 2003
Dedicated to Professor Larry Overman on the occasion of his 60th birthday.
sented by analog 3. While structurally similar to 1 and 2,
several features of this molecule merit comment.
Abstract: A new method for the synthesis of 3-alkyl-3-amino-2-
oxindoles is reported. These compounds are prepared in a 3-step
procedure using a base-mediated nucleophilic addition of benz-
ylidene-imine protected a-aminoesters to 2-nitrofluorobenzene as
the key step. The process provides a variety of 3-alkyl-3-amino-2-
oxindole analogs in yields of 1–24%. Yields are highest with ala-
nine, phenylalanine and 2-pyridylalanine as the amino acid starting
materials, while 3-pyridylalanine and O-methyltyrosine are less ef-
ficiently arylated. Sterically hindered amino acids such as valine
and phenylglycine are for all practical purposes, not substrates for
the key nucleophilic substitution reaction. The resulting 3-alkyl-3-
amino-2-oxindoles are important intermediates for the preparation
of drug-like substances. The conversion of alanine ethyl ester to 3-
amino-3-methyl-2-oxindole is described.
Compound 3 is a 3-acylamino-2-oxindole derivative, with
an enol tautomer that possesses an extremely electron-rich
aromatic ring. In fact, amine precursor 4 is extremely un-
stable as a free base (Scheme 1), decomposing almost in-
stantaneously upon formation from the corresponding
hydrochloride salt.3 Therefore, prior to initiating an ana-
log program we were compelled to address the inherent
instability of this ring. Our proposal was to quaternize the
C-3 carbon thus removing the aromatic character of the 5-
membered ring. This approach allowed us to maintain the
desirable attributes of the core, namely the drug-like prop-
erties and structural diversity afforded by the oxindole nu-
cleus while dramatically increasing the structural integrity
of our final analogs.
Key words: alkaloid, amino acid, arylation, nucleophilic aromatic
substitution, steric hindrance, oxindole
While the numerous approaches to oxindoles are avail-
able,4–11 the procedures employed limit the diversity of
final analogs. They also suffer from the lengthy synthesis
of key intermediates. For example, oxidation of indoles
provides the corresponding 2-oxindole derivatives, which
can be further elaborated into 3-alkyl-3-azido oxindoles in
a of total 4 steps.4 Direct bromination at the 3-position of
oxindoles followed by azide displacement of the resulting
The 2-oxindole moiety is a common motif in alkaloid
chemistry and is also found in a number of marketed drugs
(Figure 1).1,2 However, analogs possessing a high degree
of functionality at the 3-position are less known. As part
of a program directed at identifying potential therapies for
the treatment of obesity, we identified a series of analogs
containing a 3-substituted oxindole fragment and repre-
H
Cl
N
N
O
N
O
N
N
H
N
S
ZiprasidoneTM (1)
RopiniroleTM (2)
R
R
HN
HN
O
O
R'
R'
OH
N
O
N
H
H
3a
3b
Figure 1 Structures of oxindoles possessing clinically or biologically interesting activities.
SYNLETT 2003, No. 14, pp 2135–2138
0
6
.1
1
.2
0
0
3
Advanced online publication: 07.10.2003
DOI: 10.1055/s-2003-42074; Art ID: S06203ST
© Georg Thieme Verlag Stuttgart · New York

2136
S. J. O’Connor, Z. Liu
LETTER
amino acids other than glycine were noted, nor were re-
ports found showing this type of coupling with benzene
derivatives.
NH2-HCl
O
R
NH2-HCl
O
N
N
As shown by compound 8, we chose to substitute the
benzylidene protecting group19 for O’Donnell’s original
benzophenone-imine. Our rationale was based upon the
assumption that the smaller benzylidine group should re-
duce steric hindrance during the anion addition to fluoron-
itrobenzene. This group is also easier to add to the starting
amino acid and to remove from the product following
completion of the reaction. After some initial exploration,
we were ultimately successful in using the procedure out-
lined below (Scheme 3).
H
H
4
5
Base
<< 1 min
Base
stable
Decomposition
Scheme 1
bromide is also known.5 The latter approach necessitates
the preparation of individual oxindole starting materials,
each in turn requiring a multi-step synthesis. Al-Thebeiti
prepared 3-anilino-3-oxindole acetic acid derivatives
from isatin, but the reaction scheme limits the diversity to
acetic acid derivatives.6 Preparation of 3-spirocyclic oxin-
doles by free-radical7 and [2+3] dipolar cycloaddition8
strategies are known, but these approaches apply only to
spirocyclic analogs. The approach we ultimately settled
upon is shown retrosynthetically in Scheme 2.
R
CO2Et
R
CO2Et
a
b
N
NH2-HCl
F
Ph
NO2
10
9
R
NH2
CO2Et
R
NH2
O
c
NO2
N
H
R
R
NH2
NH2
O
11
R = CH3
12
CO2Et
N
NO2
H
Scheme 3 a) PhCHO, K2CO3, 4 Å mol. sieves, CH3CN. b) t-BuO-
Na, DMF –20 °C to r.t., then aq HCl. c) Fe, aq HCl, EtOH, reflux.
6
5
R
CO2Et
Ph
F
In practice, the chemistry is simple to perform and is rel-
atively efficient as exemplified by the conversion of ala-
nine to its corresponding oxindole derivative. Treatment
of racemic alanine methyl ester hydrochloride 9
(R = CH3) with 1.05 equivalents of benzaldehyde in the
presence of K3PO4 and 4A molecular sieves provided the
desired imine 10 in greater than 95% yield, as a single iso-
mer (1H NMR analysis, presumably the trans isomer). No
purification of this intermediate was needed. Addition of
10 to a mixture of t-BuONa and 2-fluoronitrobenzene in
anhyd DMF at –20 °C followed by slow warming to room
temperature overnight provided the crude arylated ben-
zylidene protected alanine. Other bases such as sodium
bis-trimethylsilylamide or NaOEt offered no advantages
in terms of yield or practicality. Weaker bases, such as
K2CO3 failed to provide the desired addition product. Hy-
drolysis of the benzylidene imine with HCl and extractive
work-up provided the free amine 11 in an overall yield of
41%. This material was essentially pure and could be used
without further purification.
N
+
NO2
7
8
Scheme 2
The key feature of our approach is a direct arylation of a
suitably protected amino acid derivative by nucleophilic
aromatic substitution. Though the displacement of halo-
nitro arenes by stabilized carbanions is well known,9–11
the use of amino acid derivatives as nucleophiles has re-
ceived less attention.12,13 Acetamidomalonate is known to
add to chloronitropyridines providing pyridyl glycine de-
rivatives.14 Though the reaction establishes precedent for
our strategy, the products of acetamidomalonate addition
were of limited utility for our purposes because we could
not readily convert them to more advanced intermediates.
We surmised that O’Donnell’s a-amino acid alkylation
chemistry15–17 could be extended to include arylation reac-
tions thus providing rapid access to the oxindole precur-
sors. While there are reports of O’Donnell type chemistry
succeeding in arylation reactions, these processes are lim-
ited to the use of heteroarylchlorides as coupling partners.
Even so, the yields were either very poor (8%)14 or else
were not reported.18 No examples of the direct arylation of
With a viable route to the cyclization precursor, we set out
to evaluate reduction conditions for the aryl nitro moiety.
Treatment of 11 in MeOH with 1 atmosphere of hydrogen
in the presence of 10% Pd/C provided the desired lactam,
but as an inseparable mixture of the oxindole product and
the corresponding N-hydroxylactam. The latter product
Synlett 2003, No. 14, 2135–2138 © Thieme Stuttgart · New York

LETTER
Concise Synthesis of Sterically Hindered 3-Amino-2-Oxindoles
2137
arises from the cyclization of the intermediate hydroxyl-
amine onto the ester at a rate competitive with its further
reduction to the desired aniline. The structure of the N-hy-
droxylactam was confirmed by the presence of an addi-
Table 1 Structures and 3-Step Overall Yields of Oxindoles
Prepared by Nucleophilic Addition of Protected Amino Acids to 2-
Nitrofluorobenzene
R
NH2
1
tional exchangeable proton resonance in the H NMR
O
spectrum [(d = 8.9 (s, N–OH)] as well as an observed mass
16 AMU higher than that of the desired product. Unfortu-
nately, the mixture was difficult to separate by flash chro-
matography and surprisingly resistant to further reduction
when re-subjected to the hydrogenation conditions. While
the end result was somewhat disappointing, the isolation
of the desired lactam, albeit impure validated our strategy.
N
H
12
Com- Starting
R
3-Step Yield
pound Amino Acid
13
14
Alanine
CH3
18%
24%
We immediately turned to more classical metal-mediated
reductions of nitroarenes, specifically using stoichiomet-
ric amounts of iron as the reducing agent. Thus, heating an
EtOH solution of nitro ester 11 with iron powder in the
presence of aq HCl resulted in clean reduction of the nitro
group to the corresponding aniline that under the reaction
conditions further cyclized to provide the desired oxin-
dole product in 44% isolated yield after chromatographic
purification. This is the only chromatographic purification
required during the sequence. The structure of the oxin-
dole was confirmed by its spectral properties (vida infra).
The generality of the reaction sequence was evaluated and
the results are presented in Table 1.
Phenyl-
alanine
15
16
17
2-Pyridyl-
alanine
22%
6%
N
3-Pyridyl-
alanine
N
O-Methyl-
8%
tyrosine
O
18
Tryptophan
2%
Alanine, phenylalanine and 2-pyridylalanine demonstrate
that relatively unhindered amino acids tolerate these con-
ditions. The overall yields of 18–24% reflect a per-step
yield of 56–62% on average. 3-Pyridylalanine and O-me-
thyltyrosine did not react as efficiently, providing lower
overall yields of oxindole products. Tryptophan is at best,
a poor substrate, presumably due to the relatively acidic
indole NH proton. Competition for removal of the NH
proton by tert-butoxide during the enolization step could
account for the lower yields. Prior protection of the indole
could alleviate this problem, but this was not evaluated.
Not unexpectedly, arylation at more sterically congested
centers (valine and phenylglycine) leads to a total loss of
conversion. Further optimization of the reaction condi-
tions may overcome the steric hindrance problem, but this
is only speculation. However, our hypothesis regarding
the increased stability provided by the quarternery a-car-
bon was confirmed. These quarternery 3-amino-2-oxin-
doles have significantly increased air and solution
stability and thus, are more useful as medicinal chemistry
intermediates.
N
H
19
20
Phenyl-
glycine
1%
0%
Valine
of the free amines as suitable intermediates for further
elaboration. Additionally, though not evaluated, the enan-
tioselective formation of these intermediates using chiral
cinchona alkaloids as the base is conceivable.20,21
All reagents were obtained from commercial sources and used with-
out further purification. H NMR spectra were recorded on 300
1
MHz Varian Mercury-plus spectrometer. The samples were dis-
solved in deuterated solvents obtained from Cambridge Isotope
Labs, and transferred to 5mm ID Wilmad NMR tubes. The spectra
were acquired at 293 K. The chemical shifts were recorded on the
ppm scale and were referenced to the appropriate solvent signals,
(7.26 ppm for CDCl3). High pressure liquid chromatography-elec-
trospray mass spectra (LC-MS) were obtained using either a
Hewlett-Packard 1100 HPLC equipped with a quaternary pump, a
variable wavelength detector set at 254 nm, a YMC pro C-18 col-
umn (2 × 23 mm, 120A), and a Finnigan LCQ ion trap mass spec-
trometer with electrospray ionization. Spectra were scanned from
120–1200 AMU using a variable ion time according to the number
of ions in the source. Total run time was 6.5 min.
Conclusion
We describe a new and concise method to prepare 3-alkyl-
3-amino-oxindoles. The method, with a few exceptions, is
potentially general and quite useful for preparation of ster-
ically undemanding quaternary oxindoles. Though further
optimization is warranted, this route is amenable to the
preparation of synthetically useful amounts (100–500 mg)
Synlett 2003, No. 14, 2135–2138 © Thieme Stuttgart · New York

2138
S. J. O’Connor, Z. Liu
LETTER
Preparation of Tertiary 3-Amino-2-oxindoles; General Proce-
dure
Compound 16
1H NMR (300 MHz, CD3OD): d = 8.41 (dd, J = 4.8, 1.2 Hz, 2 H),
7.73 (dt, J = 8.1, 1.5 Hz, 1 H), 7.49–7.25 (m, 3 H), 7.20–7.05 (m, 2
H), 3.01 (dd, J = 6.9, 5.7 Hz, 2 H). LC–MS for C14H13N3O: 240.5
[MH+].
Step 1. Preparation of Ethyl N-Benzylidenealaninate: D,L-Ala-
nine ethyl ester hydrochloride (3.84 g, 25.00 mmol) was suspended
in MeCN (50 mL) and sequentially treated with benzaldehyde (2.80
mL, 27.50 mmol), 4 Å molecular sieves (5 g) and powdered anhyd
K3PO4 (5.31 g, 25 mmol) and the suspension was vigorously stirred
overnight at ambient temperature. The mixture was filtered through
celite and the pad was washed well with CH2Cl2. The filtrate was
concentrated in vacuo and the residue was azeotropically dried with
toluene. This residue was place under high vacuum (1 mm Hg) to
remove the last traces of benzaldehyde (as determined by 1H NMR).
1H NMR (300 MHz, CDCl3): d = 8.40, (s, 1 H), 7.77 (m, 2 H), 7.43
(m, 3 H), 4.19 (q, 1 H), 4.09 (q, 2 H), 1.38 (d, 3 H) 1.18 (t, 3 H).
Spectral data for Compound 17
1H NMR (300 MHz, DMSO-d6): d = 10.02 (s, 1 H), 7.26 (d, J = 7.2
Hz, 2 H), 7.09 (td, J = 7.5, 1.2 Hz, 1 H), 6.93 (t, J = 7.2 Hz, 1 H),
6.71 (d, J = 8.7 Hz, 2 H), 6.59 (dd, J = 8.4, 2.7 Hz, 2 H), 3.61 (s, 3
H), 2.93 (dd, J = 32.1, 12.9 Hz, 2 H). LC–MS for C16H16N2O2: 269.0
[MH+].
Acknowledgment
Step 2. Preparation of Ethyl 2-(2-Nitrophenyl)alaninate: Ethyl
N-benzylidenealaninate (410 mg, 2.0 mmol) was dissolved in anhyd
DMF (4 mL) and the solution was cooled to –20 °C with stirring.
The cold solution was sequentially treated sequentially with 2-ni-
trofluorobenzene (0.25 mL, 2.4 mmol) and t-BuONa (250 mg, 2.6
mmol). The reaction mixture was allowed to reach r.t. overnight and
quenched by pouring into 20 mL of water. The aq mixture was ex-
tracted with EtOAc (3 × 15 mL). The combined organic extracts
were washed with water, sat. aqueous NaCl, dried over Na2SO4, fil-
tered and concentrated in vacuo. The residue was diluted with 20
mL of 1,4-dioxane and treated with 10 mL of 2 N aq HCl. After the
mixture was stirred for 3 h, it was concentrated in vacuo to remove
the dioxane, and further diluted with 30 mL of water. The aq mix-
ture was extracted with 3–20 mL portions of EtOAc and the com-
bined organic layers were washed with 20 mL of water. The aq
phases were combined and the pH adjusted to ca 10 by the careful
addition of 20 mL of 2 M aq Na2CO3. The mixture was extracted
with 3–30 mL portions of CH2Cl2 and the combined organic layers
were dried over Na2SO4, filtered and concentrated in vacuo to pro-
We are grateful to Anthony Paiva, Peter Demou and Laszlo Musza
(Molecular Analysis) for analytical support. We would also like to
thank one of the referees for alerting us the use to the prior use of
O’Donnell chemistry to prepare pyridyl glycines (refs.14,18).
References
(1) Ziprasidone: Gunasekara, N. S.; Spencer, C. M.; Keating, G.
M. Drugs 2002, 62, 1217; and references cited therein.
(2) Gallagher, G. Jr.; Lavanchy, P. G.; Wilson, J. W.; Hieble, J.
P.; DeMarinis, R. M. J. Med. Chem. 1985, 28, 1533.
(3) Upon treatment with a variety of bases (tertiary amine or
inorganic), the gray-colored solution turned immediately
purple. TLC and HPLC analysis indicated that very little if
any 3-amino-2-oxindole remained.
(4) Labroo, R. B.; Labroo, V. M.; King, M. M.; Cohen, L. A. J.
Org. Chem. 1991, 56, 3637.
(5) (a) Beccalli, E. M.; Marchesini, A.; Pilati, T. Tetrahedron
1992, 48, 5359. (b) See also: Beccalli, E. M.; Marchesini,
A. Synthesis 1992, 265.
1
vide the desired compound (194 mg, 41%). H NMR (300 MHz,
CD3OD): d = 8.27 (dd, J = 7.8, 1.2 Hz, 1 H), 8.02 (dd, J = 8,1, 1.5
Hz, 1 H), 7.93 (td, J = 7.2, 1.5 Hz, 1 H), 7.84–7.79 (m, 1 H), 4.23
(q, J = 7.2Hz, 2 H), 2.11 (s, 3 H), 1.21 (t, J = 7.5 Hz, 3 H). LC–MS:
239 [MH+].
(6) Al-Thebeiti, M. S. Heterocycles 1998, 48, 145.
(7) Coulter, T.; Grigg, R.; Malone, J. F.; Sridharan, V.
Tetrahedron Lett. 1991, 32, 5417.
(8) Jones, K.; Ho, T. C.; Wilkinson, J. Tetrahedron Lett. 1995,
36, 6743.
(9) Walsh, D. A.; Moran, H. W.; Shamblee, D. A. J. Med. Chem.
1984, 27, 1379.
(10) Urban, F. J.; Breitenbach, R.; Gonyaw, D. Synth. Commun.
1996, 26, 1629.
(11) Kraynack, E. A.; Dalgard, J. E.; Gaeta, F. C. A. Tetrahedron
Lett. 1998, 39, 7679.
(12) Harris, J. M.; Bolessa, E. A.; Mendonca, A. J.; Feng, C.-C.;
Vederas, J. C. J. Chem. Soc., Perkin Trans. 1 1995, 1945.
(13) Evans, D. A.; Evrard, D. A.; Rychnovsky, S. D.; Frueth, T.;
Whittingham, W. G.; DeVries, K. M. Tetrahedron Lett.
1992, 33, 1189.
(14) Edgar, M. T.; Pettit, G. R.; Krupa, T. S. J. Org. Chem. 1979,
44, 396.
(15) O’Donnell, M. J.; Boniece, J. M.; Earp, S. E. Tetrahedron
Lett. 1978, 19, 2641.
Step 3. Preparation of 3-Amino-3-methyl-1,3-dihydro-2H-in-
dol-2-one. Ethyl 2-(2-nitrophenyl)alaninate (550 mg, 2.00 mmol)
was dissolved in 85% EtOH (30 mL) and the mixture was treated
sequentially with iron powder (1.12 g, 56 mmol) and 2 N aq HCl (1
mL). The mixture was vigorously stirred and heated to reflux. After
stirring at the reflux temperature for 2 h, the mixture was cooled to
ambient temperature and filtered through celite. The filtrate was
concentrated to remove EtOH and diluted with water. The pH was
adjusted to ca 8 with solid K2CO3, extracted with EtOAc (3 × 10
mL), the combined extracts were washed with sat. aq NaCl and
dried over Na2SO4. The solvent was removed in vacuo and the
residue was purified by flash column chromatography (SiO2, ethyl
acetate/hexanes) to provide 176 mg (44%) of the title compound as
a white solid. 1H NMR (300 MHz, CDCl3): d = 8.16 (s, 1 H), 7.39
(d, J = 7.5 Hz, 1 H), 7.24 (t, J = 7.0 Hz, 1 H), 7.07 (t, J = 7.2 Hz, 1
H), 6.89 (d, J = 7.5 Hz, 1 H), 1.48 (s, 3 H). LC–MS: 163 [MH+].
(16) O’Donnell, M. J.; Eckrich, T. M. Tetrahedron Lett. 1978, 19,
4625.
(17) O’Donnell, M. J.; Bennett, W. D.; Jacobsen, W. N.; Ma, Y.
Tetrahedron Lett. 1989, 30, 3913.
(18) 2-Pyridylmethylamine derivatives useful as fungicides:
Maloney, B. A.; Hardy, D.; Saville-Stones, E. A. PCT Int.
Appl. WO 9942447A1, 1999.
(19) Han, O.; Frey, P. A. J. Am. Chem. Soc. 1990, 112, 8982.
(20) O’Donnell, M. J.; Wu, S.; Huffman, J. C. Teterahedron
1994, 50, 4507.
Compound 14
1H NMR (300 MHz, CD3OD): d = 7.40 (dd, J = 6.0, 0.9 Hz, 1 H),
7.14 (t, J = 8.1 Hz, 1 H), 7.07–7.00 (m, 4 H), 6.84 (dd, J = 7.8, 1.8
Hz, 2 H), 6.64 (d, J = 7.5 Hz, 1 H), 3.12 (dd, J = 16.5, 12.6 Hz, 2
H). LC–MS for C15H14N2O: 239.0 [MH+].
Compound 15
1H NMR (300 MHz, DMSO-d6): d = 10.07 (s, 1 H), 8.27 (d, J = 4.8
Hz, 1 H), 7.51 (td, J = 7.8, 2.1 Hz, 1 H), 7.10–6.96 (m, 4 H), 6.81 (t,
J = 7.2 Hz, 1 H), 6.62 (d, J = 7.8 Hz, 1 H), 3.11 (dd, J = 26.1, 12.9
Hz, 2 H), 2.28 (s, 2 H). LC–MS for C14H13N3O: 240.1 [MH+].
(21) Belokon, Y. N.; Davies, R. G.; North, M. Tetrahedron Lett.
2000, 41, 7245.
Synlett 2003, No. 14, 2135–2138 © Thieme Stuttgart · New York