Nucleophilic Addition to Electron-Rich Heteroaromatics: Dearomatizing Anionic Cyclizations of Pyrrolecarboxamides

Article abstract of DOI:10.1021/ol0364071

(Matrix presented) Despite its electron-rich nature, a pyrrole ring is susceptible to intramolecular nucleophilic attack by organolithiums. The resulting dearomatizing anionic cyclization yields new 5- or 7-membered heterocyclic rings. Formation of a new 5-membered ring, by cyclization of an N-benzylpyrrolecarboxamide, is accompanied by ring opening of the original pyrrole to yield 3-aminovinylpyrrolinones. Formation of a new 7-membered ring, by cyclization of an N-allyl pyrrolecarboxamide, yields bicyclic pyrroloazepinones.

Full text of DOI:10.1021/ol0364071

ORGANIC
LETTERS
2004
Vol. 6, No. 4
609-611
Nucleophilic Addition to Electron-Rich
Heteroaromatics: Dearomatizing Anionic
Cyclizations of Pyrrolecarboxamides
,†
Jonathan Clayden,* Rachel Turnbull,† and Ivan Pinto‡
Department of Chemistry, UniVersity of Manchester, Oxford Road, Manchester M13
9PL, UK, and GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road,
SteVenage, Herts. SG1 2NY, UK
j.p.clayden@man.ac.uk
Received December 11, 2003
ABSTRACT
Despite its electron-rich nature, a pyrrole ring is susceptible to intramolecular nucleophilic attack by organolithiums. The resulting dearomatizing
anionic cyclization yields new 5- or 7-membered heterocyclic rings. Formation of a new 5-membered ring, by cyclization of an
N-benzylpyrrolecarboxamide, is accompanied by ring opening of the original pyrrole to yield 3-aminovinylpyrrolinones. Formation of a new
7-membered ring, by cyclization of an N-allyl pyrrolecarboxamide, yields bicyclic pyrroloazepinones.
Nucleophilic addition to, and substitution at, heteroaromatic
rings is well-known for the 6-membered heterocycles such
as pyridines, pyrazines, and pyrimidines.1 Nucleophilic attack
on the electron-rich family of 5-membered heterocycles is
much less well established.2 In recent years, we have reported
examples of intramolecular nucleophilic attack by organo-
lithiums on a series of aromatic rings, both electron-rich and
electron-deficient,3 and in this paper we report the result of
using pyrroles as acceptors for organolithium nucleophiles
during anionic cyclization reactions.
2a-c followed by cycloaddition with toluenesulfonylmeth-
ylisocyanide (TosMIC). We chose initially to protect the
nitrogen atom of the pyrroles 3a-c with a methyl group and
with a range of electron-withdrawing groups (Boc, Adoc,
DEB4), which we hoped would maximize the chances of
success in the cyclization (Scheme 1).
Treatment of 4e and even 4f with base (LDA or t-BuLi)
led only to deprotection of the pyrrole nitrogen, and 3a was
recovered from the product mixture. However, treatment of
4a, which bears the yet more hindered 2,2-diethylbutanoyl
(DEB4) protecting group, with 3 equiv of LDA for 3 h at 0
°C, yielded cyclization product 9a in high yield.
The pyrrolecarboxamides 4a-f were made from the acyl
chlorides 1a-c by conversion to the acrylamide derivatives
The cyclization has evidently followed the pathway shown
in Scheme 2. The expected3,5 benzylic lithiation gives 5,
† University of Manchester.
‡ GlaxoSmithKline.
(1) For a recent example, see: Comins, D. L.; Zheng, X.; Goehring, R.
R. Org. Lett. 2002, 4, 1611.
(2) For dearomatization by other methods, see: Brooks, B. L.; Gunnoe,
T. B.; Harman, W. D. Coord. Chem. ReV. 2000, 206, 3. Donohoe, T. J.;
Garg, R.; Stevenson, C. A. Tetrahedron: Asymmetry 1996, 7, 317. Donohoe,
T. J.; Headley, C. E.; Cousins, R. P. C.; Cowley, A. Org. Lett. 2003, 5,
999 and references therein.
(3) Ahmed, A.; Clayden, J.; Rowley, M. Chem. Commun. 1998, 297.
Ahmed, A.; Clayden, J.; Yasin, S. A. Chem. Commun. 1999, 231. Clayden,
J.; Menet, C. J.; Mansfield, D. J. Org. Lett. 2000, 2, 4229. Clayden, J.;
Tchabanenko, K.; Yasin, S. A.; Turnbull, M. D. Synlett 2001, 302.
(4) Beak, P.; Zajdel, W. J. J. Am. Chem. Soc. 1984, 106, 1010. Fukuda,
T.; Mine, Y.; Iwao, M. Tetrahedron 2001, 57, 975.
10.1021/ol0364071 CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/24/2004

Scheme 1. Synthesis of the Pyrrole-3-carboxamides
Scheme 3. Cyclizations of Substituted Pyrrolecarboxamides
Scheme 2. Cyclization-Ring Opening of Lithiated
Pyrrolecarboxamides
which works well in the benzamide series)6 was frustrated
by the unavoidable formation of achiral enolate 8. However,
with the chiral and enantiomerically pure amide 14, the
cyclization product 15 cannot re-form an enolate, and indeed
15 is formed fully stereospecifically from 14: a fully
substituted stereogenic center is incorporated into the ring
with >99% ee. In agreement with previous reports of similar
cyclizations,7 both the lithiation and cyclization of 14 are
stereospecific, and the intermediate organolithium must be
configurationally stable on the time scale of the cyclization.
Remarkably, it was also possible to cyclize 16 diastereo-
selectively, despite the late deprotonation/reprotonation of
the benzylic center. The amide 17 is formed as a single
diastereoisomer. The “auxiliary” R-methyl-p-methoxybenzyl
substituent of 16 must be exerting an effect by controlling
the facial selectivity of the final protonation step, and we
are currently investigating more widely the ability of similar
auxiliaries to govern the selectivity of protonation of related
pyrrolinones.
The ring-opening of the pyrrole prevents us forming the
5,5-fused systems we had envisaged from these compounds.
However, we had noted previously that the cyclization of
an N-allyl substituted naphthamide was able to generate a
7-membered ring,8 and it seemed reasonable to suppose that
a 7,5-fused ring system might be much less strained than
the 5,5-fused systems 6. N-Allyl pyrrolecarboxamides 19a-c
were made by cycloaddition of the acrylamides 18 with
TosMIC and protection with DEBCl (Scheme 4).
which cyclizes by nucleophilic attack on the 2-position of
the pyrrole. The product, the extended enolate 6, is unstable
due to the extensive unsaturation in the 5,5-fused ring system.
A decomposition pathway, which can be rationalized either
as a 5-endo-trig or an electrocyclic ring opening, relieves
the strain and generates 7. This product must undergo an
intramolecular proton transfer to yield a new extended enolate
8, which is finally protonated to yield 9. The intermediacy
of the extended enolate 8 could be demonstrated by γ-deu-
teration, which gave 10a, or R-allylation, which gave 10b.
It was important to avoid a greater excess of LDA in the
cyclization, or byproducts 11 were generated arising from
conjugate addition of LDA to the pyrrolinone ring of 7 or 9
(with 6 equiv of LDA, 11 was formed in 20% yield).
The cyclization (effectively a conjugate substitution of the
pyrrole nitrogen by the benzyllithium) was repeated using
the substituted pyrroles 4b and 4c and again good yields of
the products 9b and 9c were obtained. Changing the
protecting group to the more readily available Piv (4d) gave
poorer yields (14%). Incorporating the pyrrole into an indole
ring system 12 produced an aminophenyl-substituted product
13 in high yield (Scheme 3).
Treatment of the N-allyl pyrrolecarboxamides 19 with
LDA gave colored allyllithiums 20 which, with the exception
of 20c, cyclized over a period of 3 h at 0 °C to yield the
enolates 23. As we had hoped, these enolates turned out to
Although the cyclizations generate chiral products, an
attempt to promote asymmetric cyclizations of 4 by using a
chiral lithium amide base in the place of LDA (a strategy
(6) Clayden, J.; Menet, C. J.; Mansfield, D. J. Chem. Commun. 2002,
38.
(7) Clayden, J.; Knowles, F. E.; Menet, C. J. Tetrahedron Lett. 2003,
44, 3397. Clayden, J.; Knowles, F. E.; Menet, C. J. Synlett 2003, 1701.
The absolute stereochemistry of 15 is assumed according to this precedent.
(8) Ahmed, A.; Clayden, J.; Rowley, M. Synlett 1999, 1954.
(5) Clayden, J. Organolithiums: SelectiVity for Synthesis; Pergamon:
Oxford, 2002.
610
Org. Lett., Vol. 6, No. 4, 2004

Scheme 4. Synthesis of N-Allyl, Crotyl, and Prenyl
Scheme 5. Pyrroloazepinones by Dearomatizing Cyclization
Pyrrolecarboxamides
Figure 1. X-ray crystal structure of cis-24a.
Scheme 6. Transformations of the Products
be stable, and stereoselective alkylation yielded in every case
essentially a single diastereoisomer of the 7,5-fused pyrrolo-
azepinones 24. The cis stereochemistry at the ring junction
of cis-24d was proved by X-ray crystal structure (Figure 2).
Products 24 are bisacylenamines, and treatment of cis-
24b with aqueous acid hydrolyzed just one of the enamides
to the amide aldehyde, yielding 25 (Scheme 6). Hydrogena-
tion by contrast was selective for the double bond in the
5-membered ring, giving 26. Hydrolysis of 26 gave pyrro-
lidine 27. The ring system of 24 is new, but is related to the
pyrroloazepinone present in the natural product hinckden-
tine A.9
Acknowledgment. We are grateful to the EPSRC for a
studentship (to R.T.) and to Dr. James Raftery for determin-
ing the X-ray crystal structures of cis-24a and cis-24d.
Figure 2. X-ray crystal structure of cis-24d.
Protonation of 23 also gave mainly the cis-fused 7,5-ring
system (Figure 1 shows an X-ray crystal structure of cis-
24a), accompanied by a small amount of a trans-fused by-
product trans-24. Cyclization of N-prenyl-substituted amide
19c gave no 7,5-fused products, presumably because of steric
hindrance at the far end of the allylic system. The only prod-
uct isolated was the pyrrolinone 22 analogous to the products
9 formed from N-benzyl amides 4 (Scheme 5).
Supporting Information Available: Experimental pro-
cedures and characterization data for all new compounds;
crystallographic data for cis-24a and cis-24d. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL0364071
(9) Liu, Y.; McWhorter, W. W. J. Am. Chem. Soc. 2003, 125, 4240.
Org. Lett., Vol. 6, No. 4, 2004
611