Alkene isomerization/enamide-ene and diene metathesis for the construction of indoles, quinolines, benzofurans and chromenes with a chiral cyclopropane substituent

Article abstract of DOI:10.1039/c0ob00597e

A synthetic method for bicyclic heterocycles, such as indole, benzofuran and chromene derivatives bearing a chiral cyclopropane at the 2-position, was established using isomerization of a terminal olefin and enamide-ene or diene metathesis. This route can also be applied to chiral 2-cyclopropylquinoline synthesis (both cis and trans).

Full text of DOI:10.1039/c0ob00597e

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 1219
www.rsc.org/obc
PAPER
Alkene isomerization/enamide-ene and diene metathesis for the construction
of indoles, quinolines, benzofurans and chromenes with a chiral cyclopropane
substituent†
Takaaki Kobayashi, Mitsuhiro Arisawa* and Satoshi Shuto*
Received 18th August 2010, Accepted 15th November 2010
DOI: 10.1039/c0ob00597e
A synthetic method for bicyclic heterocycles, such as indole, benzofuran and chromene derivatives
bearing a chiral cyclopropane at the 2-position, was established using isomerization of a terminal olefin
and enamide-ene or diene metathesis. This route can also be applied to chiral 2-cyclopropylquinoline
synthesis (both cis and trans).
Introduction
An asymmetric center adjacent to heterocyclic structures is a
motif shared by various biologically active, natural and unnatural
compounds. In many cases, the lack of an appropriate synthetic
method renders the construction of such asymmetric centers a
challenge.1
The chemistry of indole is one of the most active areas of
heterocyclic chemistry. Indole research remains at the forefront
of biological and medicinal chemistry. The most ubiquitous of the
bioactive alkaloids known are based on the indole nucleus.2 2-
Substituted indoles are also potential intermediates for synthesiz-
ing many alkaloids and pharmacologically important substances.3
While the methods for the preparation of 3-substituted indoles
Fig. 1 Our designed compound.
are well established, there is still need for yet easier access
to 2-substituted indoles. In this regard, the construction of
an asymmetric center adjacent to the 2-position of an indole
continues to be a challenging topic.4
enamide-ene ring-closing metathesis (RCM) as the key strategy.
The method was also effectively applied to the preparation of other
chiral cyclopropyl heterocycles, such as quinoline, benzofuran and
chromene derivatives.
In our work on medicinal chemistry with cyclopropane as a key
conformationally restricted unit,5 potent H3 and/or H4 receptor
antagonists 1 with a low nanomolar Ki were successfully identified
(Fig. 1).6 In the course of our studies on H3/H4 receptor ligands,
we designed the chiral 2-cyclopropylindole derivatives 2a and 2b.
However, when we started this project, there was no report on
the synthesis of indole derivatives bearing a chiral cyclopropane
at the 2-position. Therefore, we decided to establish an efficient
synthetic method for this kind of chiral indole derivative. We report
here one of the first synthetic examples of indoles with a chiral
cyclopropane at the 2-position using alkene isomerization and
Results and discussion
Synthesis of chiral 2-trans-cyclopropylindole derivatives
In 2002, we reported an alkene isomerization/enamide-ene RCM
to provide indoles effectively.7 An allylic amide 3 was isomerized
to the corresponding enamide 4, the RCM of which provided
the corresponding indole 5 in excellent overall yield (Scheme 1,
for example, from 3Aa to 5Aa). This method is applicable to the
synthesis of 2-substituted indoles 5Ab (R = Me), 5Ac (R = Et)
and 5Ad (R = Ph). However, it was difficult to synthesize those
indole derivatives having a bulky substituent, such as iso-propyl
or iso-butyl groups, because the first step, the isomerization of 3Ae
(R = iPr) and 3Af (R = iBu), did not proceed at all (Scheme 1).8
We then examined the Ru catalyst-induced isomerization of diene
3Ag, having a chiral cyclopropane substituent, which was obtained
Faculty of Pharmaceutical Sciences, Hokkaido University, Kita 12, Nishi 6,
Kita-ku, Sapporo 060-0812, Japanshu. E-mail: shu@pharm.hokudai.ac.jp,
arisawa@pharm.hokudai.ac.jp
† Electronic supplementary information (ESI) available: Experimental
details, and 1H and 13C NMR spectra of compounds 2a, 2b, 5Bf, 5Bg,
5Bh, 5De, 5Df, 5Dh, 5Di, 6e, 6f, 6g, 6h, 6i, 6j, 7, 8, 9, 10, 11, 12e, 12f,
12h, 12i, 12j, 13a, 13b, 13c, 14a, 14b, 14c, 15a, 15b and 15c. See DOI:
10.1039/c0ob00597e
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1219–1224 | 1219
©

View Article Online
allow the isomerization to occur. Therefore, focusing on the syn-
thesis of 2, we decided to continue our experiments by changing the
nitrogen substituent to a less hindered one and planned to prepare
the N-acetyl-2-ethynylaniline derivative 3Bg (Fig. 2), which has
an acetyl protecting group that is smaller than the toluenesulfonyl
(Ts) group, as a substrate for 2-substituted indole synthesis. Thus,
the establishment of a general and alternative synthetic route for
the N-allyl-2-vinylaniline derivatives 3 (Scheme 3) was needed,
since our previous method (route A in Scheme 3) had two
problems.
Scheme 1 Our previous synthesis of 2-alkyl indoles.
as a separable mixture with its regioisomer 3Ag¢. However, the
isomerization of 3Ag likewise did not proceed at all (Scheme 2).
Fig. 2 Structure of 3Bg.
(1) Substrates 3 were sometimes obtained as a mixture of
regioisomers by the Mitsunobu reaction of the corresponding N-
p-toluenesulfonyl-2-ethenylaniline and allyl alcohol. For example,
the inseparable 3Ac and SN2¢ product N-pent-2-en-1-yl-N-p-
toluenesulfonyl-2-ethenylaniline, 3Ac¢, were isolated in a ratio of
7 : 3 (Scheme 4).
Scheme 4 Our previous preparation method of 3Ac.
(2) In order for the Mitsunobu reaction to proceed, a dis-
sociative NH on 2-ethenylaniline was required. Thus, while the
highly electron-withdrawing sulfonyl groups were suitable nitrogen
protecting groups, the acetyl was not. Hence, with our previous
method toward 3Bg (Fig. 2), after the Mitsunobu reaction,
removal of the sulfonyl group, such as 4-nitrobenzensulfonyl (Ns),
and subsequent reprotection of the amino group by an acetyl
Scheme 2 Synthesis and reaction of 3Ag.
Based on these results, we considered that the steric demand of
the substrates might pose an obstacle to the initial isomerization,
namely the terminal olefin reaction site is too sterically crowded to
Scheme 3 General and alternative synthesis plan of 3.
1220 | Org. Biomol. Chem., 2011, 9, 1219–1224 This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 5 Synthesis of 2a and 2b.
group was still needed, which resulted in more steps and lower
Scope and limitation of the 2-substituted indole synthesis
yields (Scheme 3, route A).
As far as we know, this is a novel example of a chiral 2-
cyclopropylindole synthesis.4d Hence, we carried out experiments
to investigate the scope and limitation of the indole synthesis
method. First, in order to examine the effects of the protecting
group on the nitrogen and also the substituent on the alpha
position of nitrogen, we prepared the substrates 3Be, f, h (Pg =
Ac), 3Ce (Pg = Bn), and 3De, f, h–j (Pg = CHO) (Table 1) from the
corresponding precursors, the N-allyl-2-vinylaniline derivatives 6e
(R = iPr), 6f (R = iBu), 6h (R = cPr), 6i (R = chiral trans-cPr)
and 6j (R = chiral cis-cPr), in good to excellent yields. These
substrates were subsequently subjected to our synthetic method for
2-substituted indoles, and the results are summarized in Table 1.
We attempted the isomerization of 3Be (Pg = Ac, R = iPr)
using RuH(CO)Cl(PPh3)3.10 However, 3Be was recovered and none
of the desired enamide was obtained (Table 1, entry 1). The
substrate 3Ce (Pg = Bn, R = iPr) was also barely isomerized
to the corresponding enamine (entry 2) under conditions using
20 mol% of ruthenium hydride. We found that 3De (Pg = CHO,
R = iPr) was isomerized to the corresponding enamide in 23%
yield and, when 1 equivalent of RuH(CO)Cl(PPh3)3 was used, 3De
was effectively isomerized to 4De, which was further converted
to 5De in quantitative yield (3 steps, entry 3). In contrast, the
isomerization of 3Bf and 3Df proceeded (entries 4 and 5) using
20 mol% of RuH(CO)Cl(PPh3)3, and subsequent enamide-ene
RCM of the isobutyl substituted substrates 4Bf (Pg = Ac, R =
iBu) and 4Df (Pg = CHO, R = iBu) provided the corresponding
Therefore, we planned to establish an alternative prepara-
tive method for the diene substrate, including 3Bg, for the
isomerization/enamide-ene metathesis. Thus, as shown in route B
in Scheme 3, we expected that the Grignard reaction of an imine,
which was made from 2-ethenylaniline and a chiral cyclopropyl
aldehyde, with vinylmagnesium bromide would easily lead to the
diene 6 without a protecting group on the aromatic amine, where
a regioisomer could not be produced. The aromatic nitrogen
on 6 could be easily protected by the acetyl or other protecting
groups.
The reaction of 4-chloro-2-vinylaniline and the chiral trans-
9
˚
cyclopropyl aldehyde with molecular sieve pore size 4 A (MS
4A) in Et2O gave the desired imine, which was subjected to the
Grignard reaction without purification. When the reaction was
performed with vinylmagnesium bromide in Et2O in the presence
of BF3·Et2O, the desired product, 6g, was produced in 69% yield
(2 steps). Upon treatment with acetic anhydride and Hu¨nig’s
base, 6g was successfully converted to N-acetyl substrate 3Bg for
the indole ring-closure. When subjected to isomerization using
RuH(CO)Cl(PPh3)3 in refluxing xylene and a subsequent enamide-
ene RCM with 2nd generation Grubbs catalyst in refluxing
toluene, 3Bg gave the desired indole 5Bg, bearing a chiral trans-
cyclopropane at the 2-position, in 62% yield (3 steps). Compound
5Bg was successfully converted to target molecules 2a and 2b in
45% (6 steps) and 32% yield (8 steps), respectively, as shown in
Scheme 5.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1219–1224 | 1221
©

View Article Online
Table 1 Synthesis of 2-alkyl indoles
3
Entry
R
6
Method from 6 to 3
Pg
Yield of 5 (%, 3 steps)
1
2
3
4
5
6
7
8
iPr
iPr
iPr
iBu
iBu
cPr
cPr
6e
I
3Be
3Ce
3De
3Bf
3Df
3Bh
3Dh
3Di
Ac
Bn
CHO
Ac
CHO
Ac
0b
II
III
I
III
I
III
III
0b
23, quant.c
6f
6h
6i
26d
94
70
93
74
CHO
CHO
9
6j
III
3Dj
CHO
0b
a Method I: Ac2O, iPr2NEt, toluene, reflux; method II: BnBr, iPr2NEt·DMF, 100 ◦C; method III: HCOOH, Ac2O, THF, 0–60 ◦C; b The transformation
of 3 to 4 did not proceed. c In the isomerization step from 3 to 4, 1 equiv. of RuH(CO)Cl(PPH3)3 was used in refluxing toluene. d 4Bd was isolated and
RCM was performed in refluxing toluene.
indoles 5Bf and 5Df in 26 and 94% yield, respectively (3 steps). On
the other hand, the isomerization of 3Af (Pg = Ts, R = iBu) did
not proceed at all (Scheme 1), even though, like 4Bf and 4Df, it
has an iBu substituent. These results support our hypothesis that,
in sterically hindered substrates, the isomerization appears not to
proceed. The isomerization of 3Bh (Pg = Ac, R = cPr) and 3Dh
(Pg = CHO, R = cPr) also proceeded as expected. The obtained
4Bh and 4Dh were successfully converted to indoles 5Bh and 5Dh
bearing a cyclopropyl substituent at the 2-position by enamide-ene
RCM in 70 and 93% yield, respectively (entries 6 and 7).
Thus, we have succeeded in synthesizing 2-cyclopropyl indoles
5Bh and 5Dh, and isopropyl indole 5De, using a combination
of isomerization and enamide-ene RCM. This method was next
applied to the chiral 2-trans and 2-cis cyclopropylindole syntheses.
As shown in Table 1, a chiral 2-trans cyclopropylindole, 5Di, was
prepared in good yield (entry 8), although the corresponding
2-cis-cyclopropylindole derivative, 5Dj, was not obtained (entry
9), probably due to the significant steric hindrance around
the allyl moiety bearing the cis-cyclopropane structure in the
isomerization.
quinoline derivatives were previously prepared by a Suzuki–
Miyaura coupling of the trans-cyclopropylboronic acids and 2-
[(trifluoromethylsulfonyl)oxy]quinoline.11 However, no synthetic
examples of chiral 2-cis-cyclopropyl quinoline derivatives have
been reported so far. Furthermore, these methods are useful for
preparing chiral 2-cyclopropyl oxygen-containing heterocycles,
namely chromene and benzofuran derivatives (Scheme 6), the
syntheses of which have not been reported either.
Conclusion
We have developed a new method for the preparation of chiral
2-trans-cyclopropyl indoles using alkene isomerization/enamide-
ene RCM, which successfully led to our H4 antagonist candidates.
Moreover, this method is applicable to the synthesis of chiral 2-
cyclopropyl quinolines, benzofurans, and chromene derivatives
(both cis and trans).
Experimental section
General procedure for a,x-diene (6) synthesis. A solution of
aldehyde, aniline and MS4A in Et2O (SM 0.2 M) was stirred at rt
for 10 h. After the addition of Et2O (SM 0.1 M), to the resulting
Synthesis of quinoline, benzofuran and chromene derivatives
Next, we applied the above method to construct other chiral cy-
clopropyl heterocycles, e.g., quinoline, benzofuran and chromene
derivatives. The structural diversity and biological importance of
these heterocycles, found in various natural products, have made
them attractive targets for synthesis.
As shown in Table 2, 2-alkyl quinolines (12e, f, h), and also
chiral cis-cyclopropyl and trans-cyclopropyl quinolines (12i and
j), were obtained by the diene RCM of 3De, 3Df, 3Dh, 3Di and
3Dj in good to excellent yields. The chiral 2-trans-cyclopropyl
◦
solution was added BF3·Et2O at -40 C and the mixture stirred
at the same temperature for 30 min. After the addition of vinyl
magnesium bromide (1.0 M in THF) or vinyl magnesium chloride
(1.46 M in THF) at -40 ◦C, the mixture was heated gradually to
0 ◦C and stirred for 15 h. After the addition of sat. aq. NH4Cl, the
mixture was partitioned between AcOEt and sat. aq. NH4Cl. The
organic layers were washed with sat. aq. NaHCO3 and brine, dried
over Na2SO4, and concentrated at reduced pressure. The residue
1222 | Org. Biomol. Chem., 2011, 9, 1219–1224
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 2 Synthesis of 2-alkyl quinolines
Entry
R
6
Grubbs (II) (mol%)
Solvent
Yield of 12 (%, 3 step)
1
2
3
4
iPr
iBu
cPr
6e
6f
6h
6i
10
10
10
20
CH2Cl2
CH2Cl2
CH2Cl2
benzene
76
87
77
67a
5
6j
20
benzene
67a
a The corresponding TBDPS was removed and a hydroxy compound obtained.
5.31–5.12 (3H, m), 4.00 (1H, brs), 3.80–3.72 (1H, m), 3.41–3.21
(2H, m), 1.11–1.01 (10H, m), 0.98–0.92 (1H, m) and 0.54–0.41
(2H, m); 13C NMR (100 MHz, CDCl3): d 143.09, 138.24, 137.87,
135.57, 135.55, 133.73, 133.69, 133.67, 131.67, 131.53, 129.63,
129.62, 128.19, 128.11, 127.66, 127.62, 126.77, 126.67, 125.56,
125.46, 121.96, 121.90, 117.33, 115.64, 115.43, 113.25, 113.20,
66.75, 66.48, 59.60, 59.00, 26.91, 26.86, 22.54, 22.11, 19.48, 19.22,
19.20, 18.57, 7.86 and 7.67; LR-MS (FAB): m/z 501 (M+); anal.
calc. for C31H36ClNOSi: C, 74.15; H, 7.23; N, 2.79; found: C,
73.86; H, 7.33; N, 2.97%.
General procedure for N-acetyl 2-alkyl indole derivative (5B)
synthesis.
Acetylation. A solution of a,w-diene (6), N,N-diisopropyl-
ethylamine (10 equiv.) and acetic anhydride (10 equiv.) in toluene
(SM 0.1 M) was refluxed for 24 h. After cooling and addition of
sat. aq. NaHCO3, the reaction mixture was partitioned between
AcOEt and sat. aq. NaHCO3. The organic layers were washed with
sat. aq. NH4Cl and brine, dried over Na2SO4, and concentrated
under reduced pressure. The residue was purified by silica gel
column chromatography to give N-acetyl derivative (3B) as a
yellow oil.
Scheme 6 Synthesis of 2-alkyl chromene and benzofuran derivatives.
Isomerization and RCM. The mixture of the corresponding
3B and Ru(CO)HCl(PPh3)3 (20 mol%) in xylene (SM 0.1 M)
was refluxed for 24 h. After cooling, Grubbs’ 2nd generation
catalyst (20 mol%) was added and the resulting mixture stirred at
120 ◦C for 5 h. After cooling, the solvent was concentrated under
reduced pressure. The residue was purified by silica gel column
chromatography to give the N-acetylindole derivative (5B) as a
colorless oil.
was purified by Flash silica gel column chromatography (hexane)
to give the a,w-diene as a pale yellow oil.
(1R,2R)-2-t-Butyldiphenylsilyloxymethyl-1-[1-(4-chloro-2-vinyl-
phenylamino)prop-2-en-1-yl]cyclopropane (6g). 6g (35 mg,
69.3 mmol, 2 steps 69%, dr = 1 : 1) was prepared from (1R, 2R)-
2-(tert-butyldiphenylsilyloxy)methyl-1-formyl
cyclopropane5a
(34 mg, 99.8 mmol), 2-vinyl-4-chloroaniline12 (16 mg, 102 mmol),
MS4A (33 mg), BF3·Et2O (60 mL, 0.486 mmol) and vinyl
magnesium bromide (1.00 M in THF, 230 mL, 0.230 mmol).
[a]2D1 -12.7 (c 1.00, CHCl3); 1H NMR (400 MHz, CDCl3,
diastereomeric mixture): d 7.68–7.66 (4H, m), 7.44–7.34 (6H,
m), 7.22–7.20 (1H, m), 7.04–7.02 (1H, m), 6.73–6.64 (1H,
m), 6.48–6.44 (1H, m), 5.81–5.73 (1H, m), 5.63–5.56 (1H, m),
Acknowledgements
This research was supported by Grants-in-Aid from the Ministry
of Education, Culture, Sports, Science and Technology, Japan. We
also thank Sanyo Fine Co. Ltd. for providing chiral epichlorohy-
drin.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1219–1224 | 1223
©

View Article Online
Yamada, R. Kimura, S. Matsumoto, K. Furuichi, A. Matsuda and S.
Shuto, J. Med. Chem., 2003, 46, 1980–1988.
6 (a) M. Watanabe, Y. Kazuta, H. Hayashi, S. Yamada, A. Matsuda and
S. Shuto, J. Med. Chem., 2006, 49, 5587–5596; (b) M. Watanabe, T.
Hirokawa, T. Kobayashi, A. Yoshida, Y. Ito, S. Yamada, N. Orimoto,
Y. Yamasaki, M. Arisawa and S. Shuto, J. Med. Chem., 2010, 53, 3585–
3593.
Notes and references
1 T. Shinohara and K. Suzuki, Tetrahedron Lett., 2002, 43, 6937–6940.
2 (a) G. W. Gribble, in Comprehensive Heterocyclic Chemistry, Pergamon,
New York, 2nd edn, 1996, vol. 2, pp. 203–257; (b) V. Snieckus, The
Alkaloids, Academic Press, New York, 1968, vol. 11; (c) R. Gibe and
M. A. Kerr, J. Org. Chem., 2002, 67, 6247.
3 (a) M. E. Kuehne, D. E. Podhorez, T. Mulamba and W. G. Bornmann,
J. Org. Chem., 1987, 52, 347; (b) C. Hashimoto and H.-P. Husson,
Tetrahedron Lett., 1988, 29, 4563; (c) P. Leon, C. Garbay-Jaureguiberry,
M. C. Barsi, J. B. Le Pecq and B. P. Roques, J. Med. Chem., 1987,
30, 2074; (d) S. P. Modi, A.-H. Zayed and S. Archer, J. Chem. Soc.
D, 1970, 1095; (e) R. J. Sundberg, in The Chemistry of Indoles, ed.
A. T. Blomquist, Academic Press, New York, 1970.
4 (a) S. Lee and D. W. C. MacMillan, J. Am. Chem. Soc., 2007, 129,
15438–15439; (b) D. A. Evans and K. R. Fandrick, Org. Lett., 2006,
8, 2249–2252; (c) D. M. Tellers, J. C. McWilliams, G. Humphrey, M.
Journet, L. DiMichele, J. Hinksmon, A. E. Mckeown, T. Rosner, Y. Sun
and R. D. Tillyer, J. Am. Chem. Soc., 2006, 128, 17063–17073; (d) A. B.
Charette, C. Molinaro and C. Brochu, J. Am. Chem. Soc., 2001, 123,
12168–12175.
7 M. Arisawa, Y. Terada, M. Nakagawa and A. Nishida, Angew. Chem.,
Int. Ed., 2002, 41, 4732–4734.
8 Y. Kasaya, K. Hoshi, Y. Terada, A. Nishida, S. Shuto and M. Arisawa,
Eur. J. Org. Chem., 2009, 4606–4613.
9 (a) Y. Kazuta, A. Matsuda and S. Shuto, J. Org. Chem., 2002, 67, 1669–
1677; (b) Y. Kazuta, K. Hirano, K. Natsume, S. Yamada, R. Kimura,
S. Matsumoto, K. Furuichi, A. Matsuda and S. Shuto, J. Med. Chem.,
2003, 46, 1980–1988.
10 Note: Reactions using RhH(CO)(PPh3)3 (S. Krompiec, M. Pigulla,
T. Bieg, W. Szczepankiewicz, N. Kuznik, K. Leszczynska-Sejda, M.
Kubicki and T. Borowiak, Tetrahedron Lett., 2001, 42, 7095–7098) or
Fe(CO)5 (; S. Sergeyev and M. Hesse, Synlett, 2002, 8, 1313–1317) were
not successful.
11 M. L. Yao and M. Z. Deng, New J. Chem., 2000, 24, 425–428.
12 B. S. Lee, J. H. Lee and D. Y. Chi, J. Org. Chem., 2002, 67, 7884–
7886.
5 For examples, see: (a) Y. Kazuta, A. Matsuda and S. Shuto, J. Org.
Chem., 2002, 67, 1669–1677; (b) Y. Kazuta, K. Hirano, K. Natsume, S.
1224 | Org. Biomol. Chem., 2011, 9, 1219–1224
This journal is
The Royal Society of Chemistry 2011
©