Highly chemoselective reduction using a Rh/C-Fe(OAc)2 system: Practical synthesis of functionalized indoles

Article abstract of DOI:10.1016/j.tetlet.2005.11.145

Here, we report a highly effective and chemoselective method of preparing substituted indoles from (E)-2-nitropyrrolidinostyrenes via hydrogenation in the presence of a rhodium catalyst doped by additives such as Ni(NO ₃)₂·6H₂O, Fe(OAc)₂ or Co(acac)₃. These hydrogenation conditions may also be applied to other substrates. Aromatic nitro compounds and olefins can be selectively reduced in the presence of aromatic benzyl ethers, aromatic halides and aromatic aldehydes.

Full text of DOI:10.1016/j.tetlet.2005.11.145

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 969–972
Highly chemoselective reduction using a Rh/C–Fe(OAc)2 system:
practical synthesis of functionalized indoles
Atsushi Akao,a,* Kimihiko Sato,a Nobuaki Nonoyama,a Toshiaki Masea
and Nobuyoshi Yasudab
aProcess Research, Banyu Pharmaceutical Co. Ltd, 9-1, Kamimutsuna 3-Chome, Okazaki, Aichi 444-0858, Japan
bMerck Research Laboratories, Department of Process Research, PO Box 2000, Rahway, NJ 07065, USA
Received 28 October 2005; revised 24 November 2005; accepted 28 November 2005
Available online 20 December 2005
Abstract—Here, we report a highly effective and chemoselective method of preparing substituted indoles from (E)-2-nitropyrrol-
idinostyrenes via hydrogenation in the presence of a rhodium catalyst doped by additives such as Ni(NO3)2Æ6H2O, Fe(OAc)2 or
Co(acac)3. These hydrogenation conditions may also be applied to other substrates. Aromatic nitro compounds and olefins can
be selectively reduced in the presence of aromatic benzyl ethers, aromatic halides and aromatic aldehydes.
Ó 2005 Elsevier Ltd. All rights reserved.
The development of DNA topoisomerase I inhibitors
as cancer chemotherapy agents is an active area of
research.1 In recent years, the synthesis of indolocarb-
azole-class DNA topoisomerase inhibitors such as
rebeccamycin2 and arcyriaflavins3 has attracted wide-
spread interest. We are currently interested in develop-
ing an indolocarbazole glycoside DNA topoisomerase
I inhibitor (1) for this purpose.4 One of the main issues
in its manufacture is the establishment of a practical
preparation method for its indole moiety (Fig. 1).
general methods is the reductive cyclization reaction in
which (E)-2-nitropyrrolidinostyrenes (2) are converted
to indoles (3) (Scheme 1).6
Raney nickel is commonly used as a reagent in the
reductive cyclization reaction,6 but is not suitable for
large-scale use due to its pyrophoric nature and poten-
tial problems with waste disposal. Reduction with other
metals7 may be considered as an alternative, but this
usually results in lower yields compared with Raney
nickel reduction, and requires a labour-intensive work-
up process to remove metal residues. Catalytic hydro-
genation is an alternative high-yielding general method
Various methods5 have been developed for the construc-
tion of this ring system. One of the most concise and
OH
OH
HN
H
H
N
N
N
O
O
O
O
O
O
HO
OH
HO
HO
OH
N
N
N
N
N
N
H
H
H
H
Cl
Cl
O
O
OH
OH
HO
OH
HO
OH
HO
rebeccamycin
arcyriaflavin C
1
Figure 1. Indolocarbazole drug candidates.
*
Corresponding author. Tel.: +81 564 51 5668; fax: +81 564 51 7086; e-mail: atsushi_akao@merck.com
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.11.145

970
A. Akao et al. / Tetrahedron Letters 47 (2006) 969–972
Table 1. Hydrogenation of indolesa
Entry Conditions Starting
N
R
Product
yields (%)
By-product
yields (%)
R
N
materials (%)
H
NO2
OBn
OBn
OH
2
3
N
N
H
N
H
NO2
Scheme 1. Reductive cyclization reaction from (E)-2-nitropyrrolidino-
styrene to indole.
1
2
3
4
None
Nib
7 (0)
7 (0)
7 (0)
7 (0)
8 (43)
8 (88)
8 (83)
8 (80)
9 (38)
9 (6)
9 (5)
9 (3)
Fec
Cod
which does not generate stoichiometric amounts of metal
waste and can be operated safely. However, chemoselec-
tivity problems such as over-reduction of the indole rings
and unwanted reduction of other functional groups (benz-
yl ethers, aromatic halides, etc.)8 can sometimes occur.
We therefore decided to investigate the conditions for
selective catalytic hydrogenation. Here, we report a
highly chemoselective, practical method of synthesizing
indoles (3) via catalytic hydrogenation from (E)-2-nitro-
pyrrolidinostyrenes (2) in the presence of a catalytic
amount of 5% Rh/C with nickel, iron or cobalt salts.
BnO
HO
BnO
N
N
N
H
H
NO2
5
6
7
8
None
Nib
10 (0)
10 (0)
10 (0)
10 (0)
11 (86)
11 (93)
11 (99)
11 (96)
12 (14)
12 (1)
12 (1)
12 (3)
Fec
Cod
N
N
N
H
H
NO2
OBn
OBn
OH
9
10
11
None
Nib
Fec
13 (0)
13 (0)
13 (0)
14 (48)
14 (61)
14 (61)
15 (52)
15 (0)
15 (0)
Since 6-benzyloxyindole (4) is the key intermediate for
our indolocarbazole 1, the reductive cyclization of en-
amine (5) to the desired indole 4 was the subject of inten-
sive study. As a starting point we examined common
catalysts such as Pd/C, Pt/C and Rh/C. Hydrogenation
in the presence of Pd/C or Pt/C did not produce the de-
sired indoles and resulted only in over-reduced prod-
ucts.9 With Rh/C as catalyst, hydrogenation provided
a mixture of 4 and de-benzylated indole (6). To address
the issue of over-reduction, poisoned Pd or Pt catalysts
such as Lindler catalyst8,10 or Pt–sulfide8,11 were applied,
but this mainly resulted in unwanted products.9,12 Since
6 was the only by-product obtained using the Rh/C
catalyst system, we focussed on modification of this
system. Doping Rh/C with amines and/or sulfur, which
can sometimes poison the platinum group catalysts,13
did not prevent the de-benzylation reaction. It has been
a All reactions were performed under a balloon pressure of hydrogen at
room temperature. Yields were determined by HPLC after complete
conversions.
b 3 mol % of 5% Rh/C with 20 mol % of Ni(NO3)2Æ6H2O.
c 1 mol % of 5% Rh/C with 1 mol % of Fe(OAc)2.
d 1 mol % of 5% Rh/C with 5 mol % of Co(acac)3.
e 3 mol % of 5% Rh/C.
of Fe(OAc)2 resulted in the isolation of 4 in 96% yield.
This method has become one of the cornerstones of
the manufacturing process of our drug candidate 1.
To evaluate the scope and limitations of these hydroge-
nation reactions, we conducted experiments using other
indoles with benzyloxy groups at different positions on
the indole ring. The results are summarized in Table
1.20 The desired indoles were obtained in uniformly high
yields without de-benzylation under doping conditions.
In general, the 7-substituted indoles were obtained in
lower yields than the other substituted indoles due to
instability of the 7-subsituted enamine (entries 9–11).
The results may reflect a slower reaction under doping
conditions than when Rh/C is used by itself.
2+
reported that metal cations such as Pb4+
,
Mg 14 and
10
Zn2+ exhibit poisoning effects.13,15–17 Several metal
salts were screened18 as modifiers for reduction; the suc-
cessful results are summarized in Scheme 2. It was found
that combinations of Rh/C and Ni(NO3)2Æ6H2O,
Fe(OAc)2 or Co(acac)3 showed good chemoselectivity
and afforded 4 in high yields.19 In particular, the use
15
a
Chemoselectivity among aromatic nitro groups and
other reducible functional groups was examined and
the results are summarized in Table 2. In the reactions
of 3-benzyloxynitrobenzene (16) (entries 1–4), the nitro
group was selectively reduced in the presence of the benz-
yloxy group under the doping conditions. Even when
reaction times were prolonged after completion, the for-
mation of 3-hydroxyaniline (18) under doping condi-
tions was slower. The iron-doped catalyst (1 mol % of
Rh/C) afforded the best result (92% yield) (entry 3), so
other substrates were examined extensively under iron-
doped conditions. In the reactions of 4-chloronitroben-
zene (19), the nitro group was reduced in the presence
of the chloro group under both no-doping and iron-dop-
ing conditions (entries 5 and 6), but the yield was better
when the iron additive was used. For 3-bromo- and 3-
H2
Rh/C
N
+
N
N
BnO
HO
H
H
THF, rt.
BnO
NO2
4f
af
5
Additives
no additivesb
Ni(NO3)2 6H2O
55%
90%
96%
94%
34%
1%
1%
3%
c
.
d
Fe(OAc)2
e
Co(acac)3
Scheme 2. Reductive cyclization reaction to 6-benzyloxyindole.
Reagents and conditions: (a) All reactions were performed under a
balloon pressure of hydrogen at room temperature; (b) 3 mol % of 5%
Rh/C; (c) 3 mol % of 5% Rh/C with 20 mol % of Ni(NO3)2Æ6H2O; (d)
1 mol % of 5% Rh/C with 1 mol % of Fe(OAc)2; (e) 1 mol % of 5% Rh/
C with 5 mol % of Co(acac)3; (f) yields were determined by HPLC.

A. Akao et al. / Tetrahedron Letters 47 (2006) 969–972
Table 2. Hydrogenation of aromatic nitro compounds
971
Entry
Conditions
Time (h)
Starting materials (%)
Product yields (%)
By-product yields (%)
BnO
NH2
BnO
NO2
HO
NH2
1
2
3
4
Nona
Nib
15
15
33
33
16 (0)
16 (0)
16 (0)
16 (0)
17 (76)
17 (92)
17 (92)
17 (89)
18 (11)
18 (2)
18 (1)
18 (1)
Fec
Cod
Cl
Cl
NH2
NO2
NH2
5
6
None
Fec
17
15
19 (0)
19 (0)
20 (78)
20 (89)
21 (3)
21 (4)
Br
NO2
Br
NH2
7
8
Fef
6
22 (0)
23 (86)
21 (4)
21 (5)
I
NO2
I
NH2
Feg
3.5
24 (0)
25 (86)
a 3 mol % of 5% Rh/C.
b 3 mol % of 5% Rh/C with 20 mol % of Ni(NO3)2Æ6H2O.
c 1 mol % of 5% Rh/C with 1 mol % of Fe(OAc)2.
d 1 mol % of 5% Rh/C with 5 mol % of Co(acac)3.
e 1 mol % of 5% Rh/C.
f The reaction was not completed using 1 mol % of 5% Rh/C and 1 mol % of Fe(OAc)2. 5 mol % of Rh/C and Fe(OAc)2 were used.
g The reaction was not completed using 5 mol % of 5% Rh/C and 5 mol % of Fe(OAc)2. 15 mol % of Rh/C and Fe(OAc)2 were used.
iodo-nitrobenzene (22 and 24), additional amounts of
catalyst were required to complete the reaction (entries
7 and 8).
Fe(OAc)2 was much slower than in the presence of
Rh/C alone (entries 4 and 5), which indicates that aro-
matic aldehyde groups may be tolerated under iron-
doped conditions.
Iron-doping conditions can also be utilized for selec-
tive reduction of aromatic olefins (Table 3, entry 1).
Chemoselectivity among olefins, bromo groups and
benzyloxy groups was investigated under iron-additive
conditions. It was found that the olefin could be cleanly
reduced in the presence of bromo and benzyloxy groups
(entries 2 and 3). The reduction of aldehyde groups was
also investigated. Under the same conditions, the reduc-
tion of benzaldehyde in the presence of Rh/C with
In conclusion, we have developed a highly effective and
chemoselective method of preparing substituted indoles
from (E)-2-nitropyrrolidinostyrenes via hydrogenation
in the presence of a rhodium catalyst doped by additives
such as Ni(NO3)2Æ6H2O, Fe(OAc)2 or Co(acac)3. These
hydrogenation conditions may also be applied to other
substrates. Aromatic nitro compounds and olefins can
be cleanly reduced in the presence of aromatic benzyl
Table 3. Hydrogenation of other substrates
Entry
Conditions
Fea
Time (h)
16
Starting materials (%)
Product yield (%)
By-product yield (%)
1
2
26 (0)
27 (98)
—
Br
Br
Fea
2.5
28 (0)
29 (100)
27 (0)
O
O
O
BnO
N
BnO
N
HO
N
H
H
H
3
Fea
3
30 (0)
31 (99)
32 (1)
HO
OHC
4
5
Nonb
Fea
16
16
33 (17)
33 (86)
—
—
34 (56)
34 (7)
a 1 mol % of 5% Rh/C with 1 mol % of Fe(OAc)2.
b 1 mol % of 5% Rh/C.

972
A. Akao et al. / Tetrahedron Letters 47 (2006) 969–972
8. (a) Hannoubetsu Jitsuyou Shokubai; Tamara, K., Ed.;
Kagaku Kougyou Sha: Tokyo, 1970; (b) Augastine, R. L.
Catalytic Hydrogenation; Marcel Dekker: New York,
1965; (c) Freifelder, M. Practical Catalytic Hydrogenation;
John Wiley and Sons: New York, 1971; (d) Arnold, M. R.
Ind. Eng. Chem. 1956, 48, 1629; (e) Ashmore, P. G.
Catalysis and Inhibition of Chemical Reactions; Butter-
worth, 1969.
9. Only debenzylated indole 6 was obtained under Pd
catalyst conditions; indolines were obtained as major
products under Pt catalyst conditions.
10. Lindlar, H.; Dubuis, R. Org. Synth. Coll. Vol. 7, 880.
11. Dovell, F. S.; Greenfield, H. J. Am. Chem. Soc. 1965, 87,
2767.
ethers, aromatic halides and aromatic aldehydes. This
new method has been applied to the industrial-scale pro-
duction of an intermediate for an anti-cancer drug can-
didate. We are currently investigating the roles of these
doping agents in the reaction mechanism.
Acknowledgements
We thank Dr. Takayuki Nemoto and Mr. Moriaki
Ishikawa, for assistance with mass spectrometry, and
Dr. Takashi Araki, Mr. Kazusei Komatsu and Ms.
Makiko Kanamaru, for assistance with data collection.
12. After enamine 2 was hydrogenated with Pt–sulfide in THF
at room temperature for 22 h, a 2:1 mixture of indole 4
and 6-benzyloxyindoline was obtained. The resulting
mixture was stirred under air at room temperature for
9 days and the desired indole 4 was obtained in 76% yield.
13. Maxted, E. B. Adv. Catal. 1951, 3, 129.
14. Montgomery, J. B.; Hoffmann, A. N.; Glasebrook, A. L.;
Thigpen, J. I. Ind. Eng. Chem. 1958, 50, 313.
15. Adams, R.; Cohen, F. L.; Rees, C. W. J. Am. Chem. Soc.
1927, 49, 1093.
16. Yao, H. C.; Emmett, P. H. J. Am. Chem. Soc. 1961, 83,
799.
17. Rylander, P. N.; Karpenko, I. U.S. Patent 3,253,039, May
24, 1966.
18. Lithium, magnesium, aluminium, zinc, copper, nickel,
iron and cobalt salts were screened.
19. Other nickel, iron or cobalt salts (NiBr2, Ni(acac)2, FeCl2,
FeBr2, Fe(OOCCH)2, FeCl3, CoCl2, Co(acac)2) also
afforded good yields (75–95%); Ni(NO3)2Æ6H2O, Fe(OAc)2
and Co(acac)3 were found to provide the best results.
20. General procedure for hydrogenation. To a mixture of
enamine 5 (5.00 g, 15.4 mmol), iron(II) acetate (28 mg,
0.154 mmol) and 5% Rh/C (317 mg, 0.154 mmol), tetra-
hydrofuran (THF) (100 mL) were added. The atmosphere
was replaced with N2, followed by H2 and continued at
room temperature under a balloon pressure of H2 until the
starting enamine 5 and the intermediate 6-benzyloxy-1-
hydroxyindole had disappeared. After stirring for 15 h, the
atmosphere was replaced with N2 and aqueous NH3 (ca.
14%, 20 mL) was added. The following work-up opera-
tions were carried out under N2. After stirring for 20 min,
the mixture was filtered to remove the catalyst, which was
washed with THF (50 mL). The combined mixture was
extracted with toluene (50 mL) and the organic layer was
washed with aqueous citric acid (10%, 50 g), aqueous
sodium bicarbonate (5%, 50 g) and brine (20%, 50 g). The
yield of the target indole 4 was determined by HPLC (3.30
assay g, 96% yield).
References and notes
1. Long, B. H.; Balasubramanian, B. N. Exp. Opin. Ther.
Pat. 2000, 10, 635.
2. Matson, J. A.; Claridge, C.; Bush, J. A.; Titus, J.; Bradner,
W. T.; Doyle, T. W. J. Antibiot. 1989, 42, 1547.
3. Nakanishi, S.; Matsuda, Y.; Iwahashi, K.; Kase, H. J.
Antibiot. 1986, 40, 668.
4. (a) Akao, A.; Hiraga, S.; Iida, T.; Kamatani, A.; Kawa-
saki, M.; Mase, T.; Nemoto, T.; Satake, N.; Weissman, S.
A.; Tschaen, D. M.; Rossen, K.; Petrillo, D.; Reamer, R.
A.; Volante, R. P. Tetrahedron 2001, 57, 8917; (b)
Ohkubo, M.; Nishimura, T.; Honma, T.; Nishimura, I.;
Ito, S.; Yoshinari, T.; Arakawa, H.; Suda, H.; Morishima,
H.; Nishimura, S. Bioorg. Med. Chem. Lett. 1999, 9, 3307.
5. (a) Gribble, G. W. Contemp. Org. Synth. 1994, 1, 1; (b)
Hughes, D. L. Org. Prep. Proc. Int. 1993, 25, 607; (c)
Robinson, B. The Fischer Indole Synthesis; Wiley-Inter-
science: New York, 1982; (d) Sundberg, R. J. Indoles;
Academic Press: San Diego, CA, 1996; (e) Joule, J. A.
Indole and its Derivatives. In Science of Synthesis:
Houben-Weyl Methods of Molecular Transformations;
Thomas, E. J., Ed.; George Thieme: Stuttgart, 2000;
Vol. 10, Chapter 10.13, Category 2; (f) Gribble, G. W. J.
Chem. Soc., Perkin Trans. 1 2000, 1045.
6. (a) Batcho, A. D.; Leimgruber, W. Org. Synth. Coll. Vol.
7, 34; (b) Clark, R. D.; Repke, D. B. Heterocycles 1984,
22, 195.
7. Reductions using iron: (a) Kno¨lker, H.-J.; Hartmann, K.
Synlett 1993, 755; (b) Sinhababu, A. K.; Borchardt, R. T.
J. Heterocycl. Chem. 1988, 25, 1155; Reductions using
titanium(III) chloride: Somei, M.; Inoue, S.; Tokutake, S.;
Yamada, F.; Kaneko, C. Chem. Pharm. Bull. 1981, 29,
726; Reductions using aluminium amalgam: Toste, F. D.;
Still, I. W. J. Org. Prep. Proced. Int. 1995, 27, 576.