Gold(I)-catalyzed 5-endo-dig carbocyclization of acetylenic dicarbonyl compounds

Article abstract of DOI:10.1002/anie.200460844

Cyclopentenoid structures including compounds containing vinyl iodide, 1,3-diene, and heterocyclic moieties are obtained through the gold(I)-catalyzed 5-endo-dig addition of β-dicarbonyl compounds to unactivated alkynes under neutral conditions and at room temperature (RT). Both monocyclic and bicyclic cyclopentenes can be formed in excellent yields and with good diastereoselectivity (see scheme).

Full text of DOI:10.1002/anie.200460844

Communications
Synthetic Methods
pated that a 5-endo-dig variant would allow for carbocycliza-
tion onto nonterminal alkynes. To this end, we surveyed
several cationic group 11 metal triflates as catalysts for the
cyclization of b-ketoesters 1 with nonterminal a-3’-alkynyl
substituents [Eq. (1)]. Reaction of 1 with CuI and AgI triflate
Gold(i)-Catalyzed 5-endo-dig Carbocyclization of
Acetylenic Dicarbonyl Compounds**
Steven T. Staben, Joshua J. Kennedy-Smith, and
F. Dean Toste*
The importance of cyclopentanoid natural products[1] contin-
ues to inspire the development of methods for the synthesis of
5-membered rings.[2] For example, the
Table 1: Scope of AuI-catalyzed 5-endo-dig carbocyclization.[a]
Conia–ene reaction provides an atom-
economical synthesis of methylenecy-
clopentanes by the thermal cyclization
of e-acetylenic carbonyl compounds.[3]
While the classic thermal reaction is
limited to the exocyclic cyclization
mode, group 6 metal complexes catalyze
the formal 5-endo-dig addition of b-
ketoesters[4] and silyl enol ethers[5] to
alkynes. These reactions proceed via
intermediate metal vinylidenes and
therefore are limited to substrates con-
taining a terminal acetylene moiety. The
scope and utility of this reaction would
be greatly increased if nonterminal
alkynes could be employed as electro-
philes.[6] However, while examples of
the transition-metal-catalyzed 5-endo-
dig addition of heteroatom nucleophiles
to nonterminal alkynes are common,[7–9]
this class of cyclization employing
carbon nucleophiles is rare.[10]
Entry Substrate
Time
Product
Yield
1
2
R=allyl 3
R=propargyl 5
10 min
5 min
4
6
90%
96%
3
7
5 h
8
90%
4
5
R=Et 9
R=CH2OTHP 11 45 min
5 min
10
12
83%
80%
6
13
1 h
14
88%[b]
7
8
9
R=H 15
R=Me 17
R=Ph 19
1 h
15 min
10 min
16
18
20
83%
74%
94%
10
21
2 0 h
22
80%[c]
We have recently demonstrated that
cationic gold(i) complexes catalyze the
Conia–ene reaction by a mechanism
that appears to involve formation of a
gold(i) alkyne complex.[11] Thus, we
reasoned that an endocyclic Conia–ene
reaction might be feasible with group 11
metal complexes as catalysts for alkyne
activation.[8,9] Furthermore, while the
gold(i)-catalyzed Conia–ene reaction is
limited to terminal alkynes, we antici-
11
12
23
25
6 min
6 h
24
26
99%
99%[c]
[a] Reaction conditions: see Experimental Section. [b] Together with about 15% of a 1,3-diene isomer.
[c] 2mol% [(PPh 3)Au]Cl, 2mol% AgOTf.
complexes did not produce the desired cyclopentene adduct 2.
[*] S. T. Staben, J. J. Kennedy-Smith, Prof. Dr. F. D. Toste
Department of Chemistry
On the other hand, triphenylphosphinegold(i) triflate rapidly
(10 min) converted alkyne 1 into cyclopentene 2 in 93% yield.
Notably, these reactions were run under “open-flask” con-
ditions at room temperature without the necessity for dry
solvents or inert atmosphere.
University of California
Berkeley, CA 94720 (USA)
Fax: (+1)510-643-9480
E-mail: fdtoste@uclink.berkeley.edu
[**] We gratefully acknowledge the University of California, Berkeley,
Merck Research Laboratories, and Amgen Inc. for financial support.
J.J.K.-S. thanks Eli Lilly & Co. for a graduate fellowship. The Center
for New Directions in Organic Synthesis is supported by Bristol-
Myers Squibb as a Sponsoring Member and Novartis Pharma as a
Supporting Member.
Under these experimentally simple conditions, a wide
range of substrates underwent rapid 5-endo-dig cycloisome-
rization to give cyclopentene products (Table 1, entries 1–6).
In all cases no exo-cyclization was observed, including a
substrate (5) having the opportunity to undergo competitive
5-exo-dig cyclization onto a propargyl ester (entry 2). Varia-
tion at the ester (entries 1 and 2) and ketone (entry 3)
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5350
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200460844
Angew. Chem. Int. Ed. 2004, 43, 5350 –5352

Angewandte
Chemie
moieties is tolerated, although cyclization of aryl ketones
requires increased reaction times. The reaction is also
amenable to a wide range of alkynyl substituents including
alkyl, aryl (entry 9), vinyl (entry 6), and proton (entry 7),
although the latter appears to react more sluggishly. Impor-
tantly, the mild reaction conditions allow for the use of acid-
labile groups such as tert-butyl ester [Eq. (1)], tetrahydropyr-
anyl ether (entry 5), and tertiary propargyl ether (entry 11).
This cycloisomerization provides an alternative synthesis of
1,3-dienes often prepared by enyne metathesis.[12] For exam-
ple, 1,3-enyne 13 underwent rapid cyclization to give 1,3-
diene 14 in good yield (entry 6).
Having established the feasibility of the endocyclic
carbocyclization, we sought to apply this method to the
synthesis of bicyclic structures by a cyclopentene annula-
tion[13] onto a,b’-unsaturated b-ketoesters. Thus, conjugate
addition of allenyltriphenylstannane[14] to 27, followed by
gold(i)-catalyzed cyclization afforded cyclopentene 28 as a
single diastereomer [Eq. (2)]. This cyclopentene annulation
give cyclopentenyl iodides 32 and 34 as single diastereomers
in 93 and 76% yield, respectively [Eq. (4)].
We propose that these reactions proceed by a mechanism
involving nucleophilic addition of an enol to a gold(i) alkyne
complex (Scheme 1). Based on this mechanistic hypothesis,
one of the potential explanations for the lack of reactivity of
nonterminal alkynes in the gold-catalyzed 5-exo-dig cycliza-
tion[11] is that placement of the catalyst near an alkyl-
substituted carbon atom is sterically unfavorable. However,
in the transition state for the endocyclic reaction the gold
center is located near an alkyl-substituted carbon atom
without inhibiting the cyclization. We propose that the 5-
exo-dig Conia–ene reaction is limited to terminal alkynes
because of the development of 1,3-allylic strain in the
transition state. This strain is absent in the transition state
for the gold(i)-catalyzed 5-endo-dig cyclization allowing for
the participation of nonterminal alkynes.
can also be applied to the diastereoselective
formation of 5,5- (Table 1, entries 7–9) and 7,5-
fused (entry 10) bicyclic ring systems. Addition-
ally, bicyclo[3.2.1]octane 24 is available in excel-
lent yield from the 5-endo-dig cyclization of b-
ketoester 23 (entry 11). Lewis-basic groups, such
as a tertiary amine, are tolerated and thus the
gold-catalyzed reaction allows for the synthesis
of heterocyclic ring systems. For example, the
benzo-fused pyrrolizidine core of the mitosanes
(26)[15] is available in excellent yield from a
gold(i)-catalyzed cyclization of 3-hydroxyindole
Scheme 1. Proposed mechanism for the gold(i)-catalyzed 5-endo-dig carbocyclization.
25 (entry 12).
We have found that b-diketones[6] are also
viable nucleophiles under the optimized reaction conditions.
For example, 1 mol% triphenylphosphinegold(i) triflate rap-
idly and efficiently catalyzes the conversion of 1,3-dione 29
into cyclopentene 30 [Eq. (3)].
In conclusion, we have developed a gold(i)-catalyzed 5-
endo-dig carbocyclization of dicarbonyl compounds onto
appended alkynes. The reaction is carried out under open-
flask conditions and shows excellent tolerance for variation in
the ketone, ester, and alkyne substituents. As such, it provides
entry into a wide range of cyclopentanoid structures including
those containing 1,3-diene, vinyl iodide, and heterocyclic
moieties. These results further highlight the potential of
gold(I) complexes[17] to serve as catalysts for the formation of
carbon–carbon bonds by alkyne activation. Applications of
this strategy, including an asymmetric variant, are underway
in our laboratory and will be reported in due course.
Carbocyclization onto 1-halo-1-alkynes would provide a
facile entry into cyclopentenyl halides, however, transition-
metal-catalyzed addition reactions to alkynyl halides are
exceptionally rare.[16] We were, therefore, very pleased to find
that 1-iodoalkynes 31 and 33 underwent rapid cyclization to
Experimental Section
General synthetic procedure: To a small screw-cap scintillation vial
equipped with a magnetic stir bar and charged with a solution of the
a-3’-alkynyl-substituted b-dicarbonyl compound (~ 150 mg, 1 equiv)
Angew. Chem. Int. Ed. 2004, 43, 5350 –5352
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5351

Communications
in CH2Cl2 (0.4m) was added [Au(PPh3)]Cl (1 mol%) followed by
catalyzed cycloisomerization of 1,6-enynes: A. Ajamian, J. L.
Gleason, Org. Lett. 2003, 5, 2409 – 2411.
[11] J. J. Kennedy-Smith, S. T. Staben, F. D. Toste, J. Am. Chem. Soc.
2004, 126, 4526 – 4528.
[12] For a review of enyne metathesis, see: S. T. Diver, A. Giessert,
Chem. Rev. 2004, 104, 1317 – 1382.
[13] For a related cyclopentene annulation onto a,b-unsaturated
ketones, see: a) R. L. Danheiser, D. J. Carini, A. Basak, J. Am.
Chem. Soc. 1981, 103, 1604 – 1606; b) R. L. Danheiser, D. J.
Carini, D. M. Fink, A. Basak, Tetrahedron 1983, 39, 935 – 947.
[14] J. Haruta, K. Nishi, S. Matsuda, S. Akai, Y. Tamaura, Y. Kita, J.
Org. Chem. 1990, 55, 4853 – 4859.
[15] For examples of transition-metal-mediated approaches to this
ring system, see: a) J. Lee, J. D. Ha, J. K. Cha, J. Am. Chem. Soc.
1997, 119, 8127 – 8128; b) H. Kusama, J. Takaya, N. Iwasawa, J.
Am. Chem. Soc. 2002, 124, 11592 – 11593.
AgOTf (1 mol%). The cloudy white reaction mixture was then stirred
at room temperature and monitored periodically by thin layer
chromatography. Upon completion of the reaction, the mixture was
loaded directly onto a silica gel column and chromatographed with
the appropriate mixture of hexanes and ethyl acetate to give the
cycloisomerized products.
Received: June 2, 2004
À
Keywords: C C coupling · cyclization · cyclopentenes ·
dicarbonyl compounds · gold
.
[1] For examples, see: a) R. Noyori, M. Suzuki, Science 1993, 259,
44 – 45; b) A. Nangia, G. Prasuna, P. B. Rao, Tetrahedron 1997,
53, 14507 – 14545; c) M. H. Beale, J. L. Ward, Nat. Prod. Rep.
1998, 15, 533 – 548; d) M. Seepersaud, Y. Al-Abed, Tetrahedron
Lett. 2000, 41, 4291 – 4293.
[2] For reviews, see: a) B. M. Trost, Chem. Soc. Rev. 1982, 11, 141 –
170; b) M. Ramaiah, Synthesis 1984, 529 – 570; c) T. Hudlicky,
J. D. Price, Chem. Rev. 1989, 89, 1467 – 1486; d) L. Ghosez, Pure
Appl. Chem. 1996, 68, 15 – 22; e) M. A. Tius, Acc. Chem. Res.
2003, 36, 284 – 290; f) T. R. Rheault, M. P. Sibi, Synthesis 2003,
803 – 819.
[16] For a 5-exo-dig acetoxymercuration of a 1-iodo-1-alkyne, see:
a) G. A. Krafft, J. A. Katzenellenbogen, J. Am. Chem. Soc. 1981,
103, 5459 – 5466; see also the W-mediated reaction of 1-iodo-1-
alkynes involving a 1,2-iodide shift: b) T. Miura, N. Iwasawa, J.
Am. Chem. Soc. 2002, 124, 518 – 519; c) N. Iwasawa, T. Miura, K.
Kiyota, H. Kusama, K. Lee, P. H. Lee, Org. Lett. 2002, 4, 4463 –
4466.
[17] For reviews of Au-catalyzed reactions, see: a) G. Dyker, Angew.
Chem. 2000, 112, 4407 – 4409; Angew. Chem. Int. Ed. 2000, 39,
4237 – 4283; b) A. S. K. Hashmi, Gold Bull. 2003, 36, 3 – 9; see
also: C. Nieto-Oberhuber, M. P. Muꢀoz, E. Buꢀuel, C. Nevado,
D. J. Cꢁrdenas, A. M. Echavarren, Angew. Chem. 2004, 116,
2456 – 2460; Angew. Chem. Int. Ed. 2004, 43, 2402 – 2406.
[3] For a review, see: J. M. Conia, P. Le Perchec, Synthesis 1975,
1 – 19.
[4] F. E. McDonald, T. C. Olson, Tetrahedron Lett. 1997, 38, 7691 –
7692.
[5] N. Iwasawa, T. Miura, K. Kiyota, H. Kusama, K. Lee, P. H. Lee,
Org. Lett. 2002, 4, 4463 – 4466.
[6] A related PdII-catalyzed 6-endo-trig addition of 1,3-diones to
alkenes has been reported: a) T. Pei, R. A. Widenhoefer, J. Am.
Chem. Soc. 2001, 123, 11290 – 11291; b) H. Qian, R. A. Widen-
hoefer, J. Am. Chem. Soc. 2003, 125, 2056 – 2057.
[7] For a review, see: F. Alonso, I. P. Beletskaya, M. Yus, Chem. Rev.
2004, 104, 3079 – 3160.
[8] Au-catalyzed 5-endo-dig cyclization of O nucleophiles:
a) A. S. K. Hashmi, T. M. Frost, J. W. Bats, J. Am. Chem. Soc.
2000, 122, 11553 – 11554; b) A. S. K. Hashmi, L. Schwarz, J.-H.
Choi, T. M. Frost, Angew. Chem. 2000, 112, 2382 – 2385; Angew.
Chem. Int. Ed. 2000, 39, 2285 – 2288; c) A. S. K. Hashmi, P.
Sinha, Adv. Synth. Catal. 2004, 346, 432 – 438; of N nucleophiles:
d) Y. Fukuda, K. Utimoto, Synthesis 1991, 975 – 978; e) A.
Arcadi, S. D. Giuseppe, F. Marinelli, E. Rossi, Adv. Synth.
Catal. 2001, 343, 443 – 446; f) A. Arcadi, D. Bianchi, F. Marinelli,
Synthesis 2004, 610 – 618.
[9] Cu-catalyzed 5-endo-dig cyclization with N nucleophiles:
a) C. E. Castro, E. J. Gaughan, D. C. Owsley, J. Org. Chem.
1966, 31, 4071 – 4078; b) J. Fujiwara, Y. Fukutani, H. Sano, K.
Maruoka, H. Yamamoto, J. Am. Chem. Soc. 1983, 105, 7177 –
7179; Cu-catalyzed 5-endo-dig cyclization with O nucleophiles:
c) I. N. Houpis, W. B. Choi, P. J. Reider, A. Molina, H. Churchill,
J. Lynch, R. P. Volante, Tetrahedron Lett. 1994, 35, 9355 – 9358;
d) C. C. Bates, P. Saejueng, J. M. Murphy, D. Venkataraman,
Org. Lett. 2002, 4, 4727 – 4729; Ag-catalyzed addition of
O nucleophiles: e) J. A. Marshall, M. A. Wolf, E. M. Wallace, J.
Org. Chem. 1997, 62, 367 – 371.
[10] a) 5-endo-dig Michael addition: J. F. LavallØe, G. Berthiaume, P.
Deslongchamps, Tetrahedron Lett. 1986, 27, 5455 – 5458; b) HfIV-
catalyzed alkyne/allylsilane 5-endo-dig cycloisomerization: K.-I.
Imamura, E. Yoshikawa, V. Gevorgyan, Y. Yamamoto, J. Am.
Chem. Soc. 1998, 120, 5339 – 5340; c) RhI-catalyzed 5-endo-dig
hydroacylation: K. Tanaka, G. C. Fu, J. Am. Chem. Soc. 2001,
123, 11492 – 11493; d) radical 5-endo-dig carbocyclization: S.
Amrein, A. Studer, Chem. Commun. 2002, 1592 – 1593; e) Co-
5352
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 5350 –5352