An expedient synthesis of poly-substituted pyrroles from γ-ketonitriles via indium-mediated Barbier reaction strategy

Article abstract of DOI:10.1016/j.tetlet.2009.07.140

We developed an efficient synthetic strategy of poly-substituted pyrroles via an indium-mediated Barbier type allylation from γ-ketonitriles. Initial attack of allylindium species occurred at the nitrile group selectively to form the enamine intermediate,

Full text of DOI:10.1016/j.tetlet.2009.07.140

Tetrahedron Letters 50 (2009) 5744–5747
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An expedient synthesis of poly-substituted pyrroles from
via indium-mediated Barbier reaction strategy
c-ketonitriles
*
Sung Hwan Kim, Se Hee Kim, Ka Young Lee, Jae Nyoung Kim
Department of Chemistry and Institute of Basic Science, Chonnam National University, Gwangju 500-757, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
We developed an efficient synthetic strategy of poly-substituted pyrroles via an indium-mediated Barbier
type allylation from -ketonitriles. Initial attack of allylindium species occurred at the nitrile group selec-
tively to form the enamine intermediate, which reacted with the ketone group intramolecularly to fur-
Received 8 June 2009
Revised 27 July 2009
Accepted 28 July 2009
Available online 6 August 2009
c
nish the pyrroles.
Ó 2009 Elsevier Ltd. All rights reserved.
Dedicated to the memory of the late
Professor Eung Kul Ryu whose vision and
passion in organic and medicinal chemistry
was an inspiration for all
Keywords:
Indium
Barbier reaction
c
-Ketonitriles
Pyrroles
Allylindium reagents have been used extensively for the intro-
as 1a7 and examined the reaction with allylindium species as
shown in Scheme 2.
duction of allyl group in a Barbier type manner.1–3 Although allyl-
indium reagents can be added to many reactive functional groups
including aldehyde, ketone, and activated imine with acyl or tosyl
group,1,2 the reaction with nitrile has not been reported much ex-
cept for recent Yamamoto’s paper.3 According to the results intro-
duction of allylindium can be carried out with only nitrile
We chose 1a as a model compound based on the following
expectations: (i) the nitrile group of 1a and allylindium can pro-
duce the enamine intermediate (II) easily as reported3,4 and (ii)
the reactivity of the ketone group of 1a toward allylindium was ex-
pected low due to the steric interference of nearby phenyl substi-
tuent (vide infra). Compound 1a was prepared from desyl chloride
and methyl cyanoacetate (K2CO3, DMF, rt, 2 h) in 75% yield as a syn/
anti mixture (3:1).7,8 The reaction of 1a and allyl bromide in the
presence of indium powder (THF, reflux, 30 min) produced poly-
substituted pyrrole 2a in 55% yield as expected.8 In the reaction,
we separated lactone derivative 3a in trace amount (3%) as a syn/
anti mixture (1:1).8,9 The plausible mechanism for the formation
of 2a is depicted in Scheme 2. Allylindium species attacked the ni-
trile group first to produce enamine intermediate (II) which under-
went intramolecular condensation to form the pyrrole 2a.
Compound 3a might be produced via the sequential processes:
(i) reaction of allylindium to the ketone first to produce hydroxyes-
compounds having both an
a-hydrogen atom and an a-EWG
group.3,4
Recently we reported an efficient synthesis of diallylated d-val-
erolactam derivatives via an indium-mediated successive double
Barbier type allylations (Scheme 1).4 In the reaction, diallylated
d-valerolactam was formed via the sequential processes; (i) first
Barbier allylation of nitrile to produce enamine intermediate, (ii)
cyclization to cyclic N-acylimine derivative, and (iii) second Barbier
allylation to imine, as shown in Scheme 1.4
During the study we reasoned that poly-substituted pyrrole
derivatives could be synthesized from c-ketonitriles, if the reactiv-
ity of allylindium toward nitrile surpasses that of allylindium to
ketone. Suitably substituted pyrroles are the basic skeleton of
many biologically important substances,5,6 and numerous methods
for the synthesis of pyrrole derivatives have been investigated
ter (III), (ii) lactonization to c-butyrolactone (IV), and (iii) second
allylation at the nitrile of (IV).3,4
Encouraged by the results we prepared various
c-ketonitriles
extensively.5,6 In these contexts we prepared
c-ketonitrile such
1b–j and examined the synthesis of poly-substituted pyrrole deriv-
atives 2a–j and the results are summarized in Table 1 and Table 2.
Starting materials 1b–j were prepared as a syn/anti mixture from
* Corresponding author. Tel.: +82 62 530 3381; fax: +82 62 530 3389.
E-mail address: kimjn@chonnam.ac.kr (J.N. Kim).
methyl (or ethyl) cyanoacetate and the corresponding
a-chloro
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.07.140

S. H. Kim et al. / Tetrahedron Letters 50 (2009) 5744–5747
5745
O
O
In2I3
3
COOMe
COOMe
CN
Ph
NH2
Ph
- MeOH
Ph
N
Ph
NH
1st Barbier
Ref. 4
2nd Barbier
COOMe
COOMe
COOMe
COOMe
γ-cyanoester
Scheme 1.
H
allyl bromide
In, THF, reflux
HN
Ph
H2N
O
CN
COOMe
N
Ph
O
O
Ph
via CN attack
Ph
COOMe
Ph
COOMe
Ph
Ph
COOMe
Ph
(II)
γ-ketonitrile 1a (75%)
2a (55%)
(I)
CN
COOMe
O
O
K2CO3, DMF, rt
nd Barbier
O
O
CN
Ph
Ph
- MeOH
OH
2
(Z)
NH2
Ph
Ph
(IV)
O
Ph
COOMe
Cl
CN
Ph
Ph
(III)
Ph
3a (3%)
(syn/anti, 1:1)
Scheme 2.
Table 1
Synthesis of poly-substituted pyrroles 2a–f
(or bromo) ketone derivatives in the presence of K2CO3 in DMF at
room temperature.7
Entry
Nitrile 1a (%)
Pyrrole 2b (%)
For the substrates 1b–f, desired poly-substituted pyrroles 2b–f
were produced as the major products and we isolated them in
moderate yields (53–63%). We could not isolate the corresponding
lactone derivatives 3b–f in appreciable yields, although TLC obser-
vation implied the presence of the corresponding lactones in trace
amounts (vide infra). However, benzyl derivative 1g (entry 2 in Ta-
ble 2) showed the formation of appreciable amounts of lactone 3g
(9%) together with pyrrole 2g as the major product (52%). As we
noted above (vide supra), the increased amount of lactone 3g
might be a result of increased reactivity of the ketone group of
1g toward allylindium. When we used methyl-substituted sub-
strate 1h (entry 3 in Table 2), the yield of pyrrole 2h was reduced
to 43% and the lactone 3h was formed in an increased yield (27%,
5:1 mixture). The ratio of pyrrole/lactone was reversed for the sub-
strates 1i and 1j. In these cases, the yields of pyrroles (2i and 2j)
were low (6–19%) while those of lactones (3i and 3j) were in-
creased (38–53%). From the results, the reactivity of ketone moiety
of 1a and 1g–j toward allylindium was increased by reducing the
H
O
CN
COOMe
1a (75)
N
Ph
Ph
Ph
Ph
1
Ph
Ph
COOMe
2a (55)c
H
N
O
CN
Ph
Ph
COOEt
2
3
COOEt
1b (76)
2b (63)
Me
O
CN
H
N
COOMe
Ph
Ph
COOMe
2c (55)
Me
1c (70)
Cl
O
CN
H
N
size of the substituent at the b-position of c-ketonitrile.
COOMe
4
The stereochemistry of double bond of enamine moiety of 3a
and 3g–j might be Z as in the reported paper of Yamamoto.3 In
the 1H NMR spectrum of 3j in CDCl3, the NH2 peak was so broad
that we cannot read the chemical shift and this might be caused
by the intramolecular hydrogen bonding between NH2 and oxygen
atom of lactone. However, two broad singlets of NH2 appeared in
DMSO-d6 at d = 6.65 and 7.35 ppm.8 Important NOE results of 3j
are shown in Figure 1. NOE results (when H at 3a-position was irra-
diated) between hydrogen and allyl group at the ring junction
showed their cis-relationships. In addition, 2.5% increment of an-
other allyl group stated that the double bond geometry of enamine
is Z. Compounds 3a, 3g, and 3h were isolated as an inseparable dia-
stereomeric mixture (1:1–5:1 mixture) as shown in Table 2. It is
interesting to note that compound 2f can be aromatized easily by
DDQ (2,3-dichloro-5,6-dicyano-1,4-benzoquinone) oxidation to
produce 1H-benzo[g]indole derivative 4 in good yield (88%) as
shown in Scheme 3.
Ph
Ph
COOMe
2d (57)
Cl
1d (78)
MeO
O
CN
H
N
COOMe
1e (66)
5
6
MeO
COOMe
2e (53)
OMe
MeO
H
N
O
CN
COOMe
COOMe
2f (62)
1f (60)
a
Syn/anti mixture (1:1–6:1).
Conditions: nitrile 1 (1.0 mmol), allyl bromide (2.0 mmol), In (1.0 mmol), THF,
reflux, 30–60 min.
b
In summary, we developed an efficient synthetic strategy of
poly-substituted pyrroles via an indium-mediated Barbier-type
c
Compound 3a was isolated together (3%, see Table 2).

5746
S. H. Kim et al. / Tetrahedron Letters 50 (2009) 5744–5747
Table 2
The effect of substituent at b-position of c-ketonitrile 1
Entry
c
-Ketonitrile 1a (%)
CN attack (%)
CO attack (%)
O
CN
1
2a (55%) (entry 1 in Table 1)
3a (3)b (Scheme 2)
Ph
COOMe
1a
Ph
O
O
O
CN
COOMe
1g (71)
H
N
Ph
Ph
Ph
Ph
Ph
2
NH2
3g (9)b
COOMe
2g (52)
Ph
O
O
H
N
O
O
O
CN
Ph
Ph
Me
3
4
5
Ph
Ph
COOMe
1h (76)
NH2
Me
COOMe
2h (43)
Me
3h (27)c
H
N
CN
COOMe
1i (79)
O
Ph
Ph
NH2
3i (38)d
H
COOMe
2i (19)
O
H
O
CN
COOEt
H
NH2
O
N
COOEt
1j (70)
3j (53)d
2j (6)
a
b
c
Syn/anti mixture (2:1–3:1).
1:1 Mixture.
5:1 Mixture.
d
Single compound.
313-C00487). Spectroscopic data were obtained from the Korea Ba-
sic Science Institute, Gwangju branch.
1.0%
H
O
NH2
O
H
H
0.3%
H
References and notes
3.2%
3a
H
H
H
H
1. For the general review on indium-mediated reactions, see: (a) Auge, J.; Lubin-
Germain, N.; Uziel, J. Synthesis 2007, 1739–1764; (b) Li, C.-J.; Chan, T.-H.
Tetrahedron 1999, 55, 11149–11176; (c) Pae, A. N.; Cho, Y. S. Curr. Org. Chem.
2002, 6, 715–737.
2. For the indium-mediated Barbier type allylation of imine, acylimine,
sulfonimine and related compounds, see: (a) Vilaivan, T.; Winotapan, C.;
Banphavichit, V.; Shinada, T.; Ohfune, Y. J. Org. Chem. 2005, 70, 3464–3471;
(b) Schneider, U.; Chen, I.-H.; Kobayashi, S. Org. Lett. 2008, 10, 737–740; (c)
Kumar, H. M. S.; Anjaneyulu, S.; Reddy, E. J.; Yadav, J. S. Tetrahedron Lett. 2000,
41, 9311–9314; (d) Piao, X.; Jung, J.-K.; Kang, H.-Y. Bull. Korean Chem. Soc. 2007,
28, 139–142; (e) Ritson, D. J.; Cox, R. J.; Berge, J. Org. Biomol. Chem. 2004, 2,
1921–1933; (f) Lu, W.; Chan, T. H. J. Org. Chem. 2000, 65, 8589–8594; (g) Ranu, B.
C.; Das, A. Tetrahedron Lett. 2004, 45, 6875–6877.
H
2.5%
2.5%
Figure 1. Important NOEs of 7a-allylhexahydrobenzofuran-2-one 3j (H at 3a-
position was irradiated).
DDQ (1.5 equiv)
benzene, rt, 30 min
H
N
2f
COOMe
3. (a) Fujiwara, N.; Yamamoto, Y. Tetrahedron Lett. 1998, 39, 4729–4732; (b)
Fujiwara, N.; Yamamoto, Y. J. Org. Chem. 1999, 64, 4095–4101.
4 (88%)
4. Kim, S. H.; Lee, H. S.; Kim, K. H.; Kim, J. N. Tetrahedron Lett. 2009, 50, 1696–1698.
5. For the synthesis and biological activities of pyrrole derivatives, see: (a) Fan, H.;
Peng, J.; Hamann, M. T.; Hu, J.-F. Chem. Rev. 2008, 108, 264–287; (b) Gupton, J. T.
Top. Heterocycl. Chem. 2006, 2, 53–92; (c) Roy, A. K.; Pathak, R.; Yadav, G. P.;
Maulik, P. R.; Batra, S. Synthesis 2006, 1021–1027; (d) Singh, V.; Kanojiya, S.;
Batra, S. Tetrahedron 2006, 62, 10100–10110; (e) Bellina, F.; Rossi, R. Tetrahedron
2006, 62, 7213–7256; (f) Knight, D. W.; Sharland, C. M. Synlett 2004, 119–121;
(g) Knight, D. W.; Sharland, C. M. Synlett 2003, 2258–2260; (h) Foley, L. H.
Tetrahedron Lett. 1994, 35, 5989–5992; (i) Lee, H. S.; Kim, J. M.; Kim, J. N.
Tetrahedron Lett. 2007, 48, 4119–4122; (j) Kim, S. J.; Kim, H. S.; Kim, T. H.; Kim, J.
N. Bull. Korean Chem. Soc. 2007, 28, 1605–1608; (k) Kim, J. M.; Lee, S.; Kim, S. H.;
Lee, H. S.; Kim, J. N. Bull. Korean Chem. Soc. 2008, 29. 2215–2220 and further
references cited therein.
Scheme 3.
allylation of c-ketonitriles. The chemoselectivity between ketone
and nitrile functionalities toward allylindium species could be con-
trolled by providing some steric hindrance around the ketone
group.
Acknowledgments
This work was supported by the Korea Research Foundation
Grant funded by the Korean Government (MOEHRD, KRF-2008-
6. For the synthesis and biological activities of pyrrole-containing compounds, see:
(a) Thasana, N.; Worayuthakarn, R.; Kradanrat, P.; Hohn, E.; Young, L.;

S. H. Kim et al. / Tetrahedron Letters 50 (2009) 5744–5747
5747
Ruchirawat, S. J. Org. Chem. 2007, 72, 9379–9382; (b) Su, S.; Porco, J. A., Jr. J. Am.
Chem. Soc. 2007, 129, 7744–7745; (c) Dhawan, R.; Arndtsen, B. A. J. Am. Chem.
Soc. 2004, 126, 468–469; (d) Pandey, P. S.; Rao, T. S. Bioorg. Med. Chem. Lett. 2004,
14, 129–131.
126.38, 126.65, 128.28, 128.31, 134.35, 134.71, 137.47, 166.24; ESIMS m/z 268
[M+H]+. Anal. Calcd for C17H17NO2: C, 76.38; H, 6.41; N, 5.24. Found: C, 76.12; H,
6.65; N, 5.11.
Compound 2g: 52%; pale yellow oil; IR (film) 3304, 1676, 1465 cmÀ1 1H NMR
;
7. Prepared starting materials were identified by their IR, 1H NMR, and mass data.8
Compounds 1i and 1j were prepared according to the references, see: (a) Lev, I.
J.; Ishikawa, K.; Bhacca, N. S.; Griffin, G. W. J. Org. Chem. 1976, 41, 2654–2656; (b)
Sakai, T.; Amano, E.; Kawabata, A.; Takeda, A. J. Org. Chem. 1980, 45, 43–47.
8. Typical procedure for the synthesis of compounds 2a and 3a: A mixture of desyl
chloride (346 mg, 1.5 mmol), methyl cyanoacetate (297 mg, 3.0 mmol), and
K2CO3 (276 mg, 2.0 mmol) in DMF (3 mL) was stirred at room temperature for
2 h. After the usual aqueous workup and column chromatographic purification
process (hexanes/EtOAc, 8:1), compound 1a was obtained as colorless oil,
330 mg (75%) as a syn/anti mixture (3:1). A stirred mixture of compound 1a
(293 mg, 1.0 mmol), allyl bromide (242 mg, 2.0 mmol), and indium (115 mg,
1.0 mmol) in THF (1.5 mL) was heated to reflux for 30 min. After the usual
aqueous workup and column chromatographic purification process (hexanes/
CH2Cl2/EtOAc, 16:1:1), we obtained compound 2a (175 mg, 55%) and compound
3a (10 mg, 3%). Other compounds were synthesized similarly and the
spectroscopic data of 1a, 2a, 2f, 2g, 3a, 3j, and 4 are as follows.
(CDCl3, 300 MHz) d 3.63 (s, 3H), 3.80 (dt, J = 6.9 and 1.2 Hz, 2H), 4.21 (s, 2H),
5.19–5.26 (m, 2H), 5.94–6.07 (m, 1H), 7.10–7.14 (m, 3H), 7.20–7.36 (m, 7H), 8.31
(br s, 1H); 13C NMR (CDCl3, 75 MHz) d 31.21, 32.17, 50.36, 111.56, 118.07,
119.98, 125.24, 127.09, 127.23, 127.90, 128.04, 128.77, 129.16, 132.34, 134.27,
137.23, 142.50, 165.93; ESIMS m/z 332 [M+H]+.
Compound 3a (major/minor, 1:1): 3%; pale yellow oil; IR (film) 3466, 3411, 3326,
1694, 1625, 1556, 1239 cmÀ1 1H NMR (CDCl3, 300 MHz)
; d 2.16–2.18 (m,
2H * 0.5), 2.41–2.92 (m, 6H * 0.5), 4.20 (s, 1H * 0.5), 4.28 (s, 1H * 0.5), 4.66–5.81
(m, 12H * 0.5), 6.81–7.07 (m, 8H * 0.5), 7.25–7.45 (m, 12H * 0.5); 13C NMR
(CDCl3, 75 MHz) d 37.37, 37.40, 43.52, 47.76, 55.07, 57.17, 87.49, 88.82, 92.22,
94.51, 117.92, 118.87, 119.97, 120.19, 124.95, 126.13, 126.27, 127.00, 127.23,
127.43, 127.72, 128.31, 128.69, 128.86, 130.95, 131.14, 132.40, 132.87, 140.09,
141.08, 141.84, 145.83, 157.84, 158.10, 172.90, 173.29, and two carbon peaks
were overlapped; ESIMS m/z 346 [M+H]+.
Compound 3j: 53%; colorless oil; IR (film) 3424, 3330, 1693, 1630, 1565,
1227 cmÀ1 1H NMR (DMSO-d6, 600 MHz) d 1.08–1.53 (m, 5H), 1.73–1.79 (m,
;
Compound 1a (major/minor, 3:1): 75%; colorless oil; IR (film) 2252, 1749, 1682,
2H), 1.84–1.85 (m, 1H), 2.13–2.19 (m, 1H), 2.24–2.28 (m, 1H), 2.65 (dd, J = 8.4
and 6.0 Hz, 1H), 2.86 (dd, J = 14.4 and 6.0 Hz, 1H), 2.93 (dd, J = 14.4 and 6.6 Hz,
1H), 5.03–5.19 (m, 4H), 5.70–5.77 (m, 1H), 5.81–5.88 (m, 1H), 6.65 (br s, 1H),
7.35 (br s, 1H); 13C NMR (CDCl3, 75 MHz) d 19.25, 20.48, 29.45, 30.51, 37.10,
40.72, 45.68, 83.25, 96.63, 118.47, 119.39, 132.30, 132.92, 154.59, 173.44; ESIMS
m/z 248 [M+H]+. Anal. Calcd for C15H21NO2: C, 72.84; H, 8.56; N, 5.66. Found: C,
72.89; H, 8.77; N, 5.48.
1449, 1266 cmÀ1 1H NMR (CDCl3, 300 MHz) d 3.72 (minor, s, 3H), 3.81 (major, s,
;
3H), 4.06 (minor, d, J = 7.2 Hz, 1H), 4.56 (major, d, J = 10.2 Hz, 1H), 5.20 (major, d,
J = 10.2 Hz, 1H), 5.30 (minor, d, J = 7.2 Hz, 1H), 7.27–7.41 (major and minor, m,
14H), 7.46–7.53 (major and minor, m, 2H), 7.88–7.95 (major and minor, m, 4H);
ESIMS m/z 294 [M+H]+. Anal. Calcd for C18H15NO3: C, 73.71; H, 5.15; N, 4.78.
Found: C, 73.98; H, 5.37; N, 4.66.
Compound 2a: 55%; pale yellow solid, mp 135–136 °C; IR (KBr) 3302, 1681,
Compound 4: 88%; pale yellow solid, mp 208–210 °C; IR (KBr) 3324, 1713, 1466,
1460 cmÀ1
;
1H NMR (CDCl3, 300 MHz) d 3.60 (s, 3H), 3.82 (dt, J = 6.6 and 1.2 Hz,
1314, 1180 cmÀ1 1H NMR (CDCl3, 300 MHz) d 3.97 (s, 3H), 4.08 (dt, J = 6.6 and
;
2H), 5.22–5.31 (m, 2H), 5.97–6.11 (m, 1H), 7.06–7.31 (m, 10H), 8.39 (br s, 1H);
13C NMR (CDCl3, 75 MHz) d 32.09, 50.53, 111.92, 118.26, 123.58, 126.38, 126.65,
126.88, 127.59, 127.89, 128.44, 130.68, 132.06, 134.20, 135.67, 136.58, 165.78;
ESIMS m/z 318 [M+H]+. Anal. Calcd for C21H19NO2: C, 79.47; H, 6.03; N, 4.41.
Found: C, 79.54; H, 6.25; N, 4.31.
1.2 Hz, 2H), 5.30–5.37 (m, 2H), 6.04–6.17 (m, 1H), 7.43–7.48 (m, 1H), 7.52–7.57
(m, 1H), 7.64 (d, J = 9.0 Hz, 1H), 7.93–7.99 (m, 2H), 8.23 (d, J = 9.0 Hz, 1H), 8.99
(br s, 1H); 13C NMR (CDCl3, 75 MHz) d 32.27, 50.93, 109.72, 118.75, 119.09,
120.87, 121.02, 122.45, 123.53, 124.29, 125.77, 128.93, 129.10, 130.43, 133.80,
142.33, 166.24; ESIMS m/z 266 [M+H]+. Anal. Calcd for C17H15NO2: C, 76.96; H,
5.70; N, 5.28. Found: C, 77.07; H, 5.71; N, 5.09.
Compound 2f: 62%; pale yellow oil; IR (film) 3304, 1672, 1467 cmÀ1 1H NMR
;
(CDCl3, 300 MHz) d 2.89–3.03 (m, 4H), 3.80–3.82 (m, 2H), 3.82 (s, 3H), 5.18–5.26
(m, 2H), 5.93–6.07 (m, 1H), 7.04–7.20 (m, 4H), 8.46 (br s, 1H); 13C NMR (CDCl3,
75 MHz) d 21.46, 29.46, 32.04, 50.65, 110.18, 117.94, 118.07, 121.82, 125.55,
9. For the similar enaminobutyrolactones, see: (a) Duron, S. G.; Gin, D. Y. Org. Lett.
2001, 3, 1551–1554; (b) Bower, J. F.; Szeto, P.; Gallagher, T. Org. Lett. 2007, 9,
4909–4912.