Gold(I)/(III)-catalyzed rearrangement of divinyl ketones and acyloxyalkynyloxiranes into cyclopentenones

Article abstract of DOI:10.1021/ol403663j

Multifaceted gold(I/III) catalysts with their carbophilic and oxophilic characters catalyzed very efficiently the formation of hydroxylated cyclopentenones from simple divinyl ketones or acyloxyalkynyloxiranes. The Nazarov reaction is rapidly performed in

Full text of DOI:10.1021/ol403663j

Letter
pubs.acs.org/OrgLett
Gold(I)/(III)-Catalyzed Rearrangement of Divinyl Ketones and
Acyloxyalkynyloxiranes into Cyclopentenones
Marie Hoffmann,† Jean-Marc Weibel,† Pierre de Fremont,‡ Patrick Pale,*,† and Aurelien Blanc*,†
́
́
†Laboratoire de Synthes
7177-CNRS, Universite
̀
e, Rea
de Strasbourg, 4 rue Blaise Pascal, 67070 Strasbourg, France
́ ́
ctivite
Organiques et Catalyse and ‡Laboratoire de Chimie de Coordination, Institut de Chimie, UMR
́
S
* Supporting Information
ABSTRACT: Multifaceted gold(I/III) catalysts with their carbophilic and oxophilic characters catalyzed very efficiently the
formation of hydroxylated cyclopentenones from simple divinyl ketones or acyloxyalkynyloxiranes. The Nazarov reaction is
rapidly performed in dichloroethane with 5 mol % of the simple gold(III) trichloride salt at 70 °C, while the rearrangement of
alkynyloxiranes requires 5 mol % of a more stable NHC gold(III) triflimidate complex.
unctionalized five-membered rings, especially hydroxylated
enynyl ketones could give rise to Nazarov products, either as
Fcyclopentenones, are important units of numerous natural intermediates toward furocyclopentenones or after heterocyc-
lization.9 Since we previously synthetized acyloxylated divinyl
ketones 2 by a gold-catalyzed rearrangement of acyloxypropy-
nyloxiranes 1 (Scheme 2, step 1),10 we looked for gold-
products, such as the antibiotic tylopilusins1 isolated from
mushrooms or the anti-HIV and anti-angiogenesis terpestacin
family2 (Scheme 1). Such rings are also key building blocks
toward various useful compounds, from prostaglandin and
related drugs3 to the cyclotene family4 of fragrance ingredients
(Scheme 1).
Scheme 2. Gold(I)-Catalyzed Rearrangement of
Alkynyloxiranes into Divinyl Ketones
Scheme 1. Some Examples of Hydroxylated Cyclopentenone
Natural Products and Lewis Acid Catalyzed Nazarov
Reaction Toward the Cyclopentenone Motif
catalyzed conditions able to convert these divinyl ketones to
cyclopentenones (Scheme 2, step 2) and then envisaged a one-
pot cascade reaction in order to directly obtain five-membered
rings from alkynyl epoxides (Scheme 2, cascade). In the present
communication, we present our results demonstrating the
viability of both the Au-catalyzed Nazarov reaction and the
direct conversion of acyloxypropynyloxiranes to cyclopente-
nones in the presence of appropriate gold catalyst.
To look for the best conditions to produce cyclopentenones
from divinylketones, the simple 3-(2,2-dimethylpropionyloxy)-
2,5-dimethylhexa-1,4-dien-3-one 2a was easily prepared in 3
steps including step 1 in Scheme 2 (see Supporting
Information) and submitted to various gold catalysts (Table
1). Gold(I) derivatives, even as their electrophilic cationic
forms, did not lead to any transformation at room temperature
As a result of the various interests of hydroxylated
cyclopentenones, numerous routes have been developed to
achieve their synthesis. Among them, the Nazarov reaction
provides unique perspectives due to its stereoselectivity5 and, of
course, has been used as a key step in many natural product
syntheses.6 This reaction required divinyl ketones as substrates
and a Lewis acid as catalyst (Scheme 1). Although various
Lewis acids have already been used,7 gold salts or complexes
have surprisingly never been explored in the direct cyclization
of divinyl ketones.8 However, it is worth mentioning that some
Received: December 18, 2013
Published: January 23, 2014
© 2014 American Chemical Society
908
dx.doi.org/10.1021/ol403663j | Org. Lett. 2014, 16, 908−911

Organic Letters
Letter
Table 1. Screening of Reaction Conditions for the Au-
Catalyzed Transformation of Divinyl Ketone 2a into
Cyclopentenones 3a and 4a
Table 2. Scope of Gold(III)-Catalyzed Nazarov Reaction
entry catalyst (5 mol %) time (h) cumulative yield (%) 3a:4a (%)
b
1
PPh3AuCl/AgSbF6
PPh3AuCl/AgSbF6
PPh3AuNTf2
24
0.5
6
c
c
2
91
87
87
81
92
98
97
82
93
85
c
14.1/1
11.4/1
3.8/1
5.8/1
12.1/1
6/1
3
4
PPh3AuCl/AgOTf
4.5
16
0.5
1
d
5
L1AuSbF6.MeCN
e
6
L2AuCl/AgSbF6
7
IPrAuCl/AgSbF6
AuCl3
8
0.1
2
9.8/1
7.2/1
17.6/1
1.7/1
c
9
NaAuCl4.2H2O
It-BuAuCl2NTf2
AgSbF6
10
11
12
0.75
18
24
a
Reactions run under argon at 70 °C in DCE, c = 0.5 mol/L.
b
c
d
e
Reaction run at rt. No conversion. L1 = JohnPhos. L2 = tris(2,4-di-
tert-butylphenyl)phosphite.
(Table 1, entry 1) but readily produced the expected
cyclopentenone in 1,2-dichloroethane at 70 °C in high yields
(Table 1, entries 2−7). However, a mixture of the expected 2-
pivaloyloxyclopentenone 3a and its free form (2-hydroxy) 4a
were isolated. The former was the major one, while the latter
could usually be detected only in trace amounts (3a/4a, 6−15/
1), except when silver triflate was used to in situ prepare the
active catalyst (Table 1, entry 4), suggesting that adventitious
water would be responsible for some in situ ester hydrolysis.
Despite high overall yields, large variation in reaction time was
observed, the more electrophilic and bulkier ligands giving the
faster reactions (Table 1, entries 2 and 6 vs 3 and 5).
Interestingly, NHC-gold(I) complex combined both very high
efficiency and high reaction rate (Table 1, entry 7). Not so
surprisingly, the more electrophilic gold(III) complexes
exhibited higher reactivity, very rapidly giving 3a in high yields
(Table 1, entries 8 and 9). Among them, the simple gold
trichloride proved to be the most effective, quantitatively
yielding to the cyclized products (Table 1, entry 8). Here also,
NHC-gold(III) triflimidate complex11 exhibited an excellent
activity along with a high selectivity in favor of the pivaloylated
product 3a (Table 1, entry 10). Control experiments revealed
that silver hexafluoroantimonate could catalyze this reaction,
but after long reaction time and with a poor selectivity (Table
1, entry 11), while without catalyst, no transformation occurred
(Table 1, entry 12).
With these conditions in hand, the scope of this gold(III)-
catalyzed Nazarov reaction was then explored. Various acyloxy
divinyl ketones 2b−j were thus prepared from alkynyloxiranes
1a−j (see Supporting Information) and submitted to AuCl3 or
It-BuAuCl2NTf2 in 1,2-dichloroethane at 70 °C (Table 2). We
first examined the nature of the ester moiety in order to
improve the protected/deprotected ratio (3/4). As expected,
acetoxy and benzoyloxy divinyl ketones 2b and 2c gave the
corresponding Nazarov products in excellent yields (Table 2,
entries 1 vs 2 and 3). Interestingly, an excellent selectivity was
obtained from the latter 2c with a 20 to 1 ratio of products 3c/
4a, while a modest selectivity (3.8/1) could be achieved with
a
b
Cumulative yield (%). Reaction run with 5 mol % of AuCl3.
c
d
Reaction run with 5 mol % of It-BuAuCl2NTf2. 20% of starting
e
f
material was recovered. Ratio cis/trans. Reaction run at rt.
the former 2b, probably reflecting the better resistance toward
hydrolysis of these esters. Nevertheless, we kept the migratory
pivaloyl group for comparison due to its equal robustness
during the cascade reaction (Table 4, entries 1 vs 3). Variation
of the substitution in position R2 and R3 was then evaluated
with substrates 2d−f (Table 2, entries 1 vs 4−6). Very good
909
dx.doi.org/10.1021/ol403663j | Org. Lett. 2014, 16, 908−911

Organic Letters
Letter
reactivity and yield were maintained switching from dimethyl to
Ph/H substituents, but more hindered substituents such as
Ph(CH2)2/Me and particularly cyclopentyl clearly increased the
reaction times. In these cases, only the use of It-BuAuCl2NTf2
as catalyst restored the reactivity, leading to high to excellent
yields of the corresponding cyclopentenones, but to the
detriment of the 3/4 ratio (Table 2, entries 5 and 6). It is
worth mentioning that spiro cyclopentenones such as 3f/4f
could thus be very efficiently obtained. Bicyclopentenone
derivatives, motifs occurring in various natural products,12
could also be prepared from the corresponding cycloalkenyl
vinyl ketones 2g−i (Table 2, entries 7−9). Nevertheless, while
bicyclo[4.3.0]nonane compounds 3h/4h were nicely formed in
standard conditions, substrates 2g and 2i required the use of
NHC-gold(III) complex to reach full or correct conversions
(Table 2, entries 8 vs 7 and 9). The diastereoselectivity of the
reaction was further investigated using substrates 2j and 2k
(Table 2, entries 10 and 11). The dienone 2j rapidly furnished
the cyclopentenone 3j as mixture of isomers, with only traces of
4j using It-BuAuCl2NTf2. Despite the 4π conrotatory process,
the cis isomer was mostly produced, although in a modest ratio
(2.1/1), as determined by NOE experiment. Finally, the
dihydropyranyl vinyl ketone 2k, a benchmark substrate for the
Nazarov reaction, afforded excellent diastereoisomeric ratio in
favor of the cis isomer 5k, as usual for this compound,13 in a
quantitative yield with AuCl3 at room temperature (Table 2,
entry 11).
Having demonstrated the direct Au-catalyzed Nazarov
reaction from divinyl ketones, we then looked for a one-pot
process, starting from acyloxypropynyloxiranes 1 (Cascade in
Scheme 2). As already described by us,10 pivaloylalkynyloxirane
1a efficiently afforded divinyl ketone 2a, the transient
intermediate in route to cyclopentenones 3a/4a, in the
presence of hexafluoroantimonate triphenylphosphinogold(I)
complex at room temperature (Table 3, entry 1). Knowing that
gold-catalyzed Nazarov reaction required warm conditions, we
also evaluated the more stable triflimide gold(I) complex14 and
were pleased to isolate 2a in slightly better yield (Table 3,
entries 2 vs 1). Heating 1a for 24 h in DCE with Gagosz’s or
gold trichloride catalysts failed to afford the expected
cyclopentenones 3a or 4a, and only divinyl ketone 2a was
produced during the reaction (Table 3, entries 3 and 4 vs Table
1, entries 3 and 8). Thinking that degradation of the catalyst
could occur during the first step, we conducted the one-pot
reaction sequentially by adding another amount of catalyst (5
mol % of PPh3AuNTf2 or AuCl3) at 70 °C once 2a was formed
(Table 3, entries 5 and 6). Under such conditions, we
successfully obtained cyclopentenones in excellent cumulative
yields, although with low selectivity for 3a vs 4a. After a large
screening of catalysts, we finally found out that only 5 mol % of
dichlorotriflimidate It-Bu-gold(III) complex was able to catalyze
the cascade reaction in 76% yield (Table 3, entry 7). It is
noteworthy that the best ratio of 3a/4a was obtained by
forming 2a at room temperature and then heating the reaction
mixture at 70 °C to complete the formation of cyclopentenones
(Table 3, entries 7 vs 8).
Having established the optimum reaction conditions (Table
3, entry 7), we next evaluated the scope of this rearrangement
on various acyloxyalkynyloxiranes 1a−j (Table 4). Upon gold
Table 4. Scope of It-BuAuCl2NTf2-Catalyzed Formation of
Cyclopentenones 3:4 from Acyloxyalkynyloxiranes 1
time
a
entry
1
acyloxyalkynylepoxide
products (h)
yield (3:4)
1a R1,2,3 = Me, R4 = H,
3a/4a
2
76 (8.5/1)
R5 = t-Bu
2
1b R1,2,3 = Me, R4 = H, R5 =
3b/4a
2
57 (2.6/1)
Me
3
4
1c
R
1,2,3 = Me, R4 = H, R5 = Ph 3c/4a
1
1
75 (8.3/1)
74 (3.6/1)
1d R1,4 = H, R2 = Ph, R3 = Me, 3d/4d
R5 = t-Bu
5
1e R1,3 = Me, R2 = (CH2)2Ph,
R4 = H, R5 = t-Bu
3e/4e
3f/4f
1.5
78 (1.7/1)
71 (1.4/1)
90 (2.9/1)
89 (8.9/1)
97 (2.6/1)
b
6
1f
R1,2 = -(CH2)4-, R3 = Me,
24
2
R4 = H, R5 = t-Bu
7
1g R1,2 = Me, R3,4 = -(CH2)3-,
3g/4g
3h/4h
3i/4i
R5 = t-Bu
8
1h R1,2 = Me, R3,4 = -(CH2)4-,
R5 = t-Bu
2.5
2.5
9
1i
R1,2 = Me, R3,4 = -(CH2)5-,
R5 = t-Bu
Table 3. Screening of Reaction Conditions for the Tandem
Gold-Catalyzed Rearrangement of Acyloxyalkynyloxiranes
1a into Cyclopentenones 3a and 4a
c
10
1j
R1 = H, R2 = Ph,
3j/4j
1
80 (6.9/1)
d
R3,4 = -(CH2)5-, R5 = t-Bu
dr 2.3/1
a
b
c
Cumulative yield (%). Reaction started at 70 °C. 4j isomerized on
d
silica gel into 2-hydroxy-3-phenylcyclopent-2-enone derivative. Ratio
of cis/trans for 3j/3j′ products.
catalysis, pivaloyl and benzoyl derivatives 1a and 1c exhibited
the same excellent reactivity and yield, while the acetoxy
compound 1b clearly showed a lack of selectivity and gave
modest yield (Table 4, entries 1 and 3 vs 2) presumably due to
the less efficient formation of divinyl ketone 2b (see Supporting
Information). We then looked at the influence of R1 and R2
substitution on the cascade (Table 4, entries 4−6). As for the
Nazarov reaction (Table 2, entries 4−6), compound 1f
underwent the reaction at a considerably slower rate than 1d
and 1e, but in the all cases good yields have been maintained.
Upon cascade conditions, alkynyl epoxides 1g−1i afforded
bicyclic systems with even better yields than from divinyl
ketones (Table 4, entries 7−9). The stereoselectivity of the
cascade was investigated with substrate 1j. An equal ratio (2.3/
b
time
(h)
yield (%) yield 2a
entry catalyst (5 mol %)
temp (°C)
(3a:4a)
(%)
1
2
3
4
5
PPh3AuSbF6
PPh3AuNTf2
PPh3AuNTf2
AuCl3
0.33
0.33
24
rt
84
87
62
57
rt
rt → 70
rt → 70
24
trace
PPh3AuNTf2 then
+ 5 mol %
0.33
1
rt
70
80 (3.2/1)
6
PPh3AuNTf2 then
AuCl3
0.33
1
rt
70
87 (4.1/1)
7
8
It-BuAuCl2NTf2
It-BuAuCl2NTf2
2
1
rt → 70
76 (8.5/1)
70
70 (3.8/1)
a
b
Reactions run under argon in DCE, c = 0.5 mol/L. Cumulative yield.
910
dx.doi.org/10.1021/ol403663j | Org. Lett. 2014, 16, 908−911

Organic Letters
Letter
1) to the direct Nazarov reaction (Table 2, entry 10) was
observed in favor of the cis isomer (Table 4, entry 10). It is
noteworthy that for the large majority of substrates, the cascade
yield is better than the overall yields of the two independent
gold steps.
Based on multifaceted gold catalyst properties, i.e., the ability
of gold cations to act as π or σ Lewis acid, the following
mechanistic hypothesis could be envisaged for the rearrange-
ment of alkynylepoxides 1 into cyclopentenones 3 via transient
divinyl ketone 2 (Scheme 3). As it was previously proposed,
ACKNOWLEDGMENTS
■
We gratefully acknowledge the Agence Nationale de la
Recherche for a grant (ANR-11-JS07-001-01 Synt-Het-Au)
and the CNRS. MH thanks the French Ministry of Research for
a PhD fellowship.
REFERENCES
■
(1) Fukuda, T.; Nagai, K.; Tomoda, H. J. Nat. Prod. 2012, 75, 2228.
(2) (a) Oka, M.; Iimura, S.; Tenmyo, O.; Sawada, Y.; Sugawara, M.;
Ohkusa, N.; Yamamoto, H.; Kawano, K.; Hu, S.-L.; Fukagawa, Y.; Oki,
T. J. Antibiot. 1993, 46, 367. (b) Jung, H. J.; Lee, H. B.; Kim, C. J.;
Rho, J.-R.; Shin, J.; Kwon, H. J. J. Antibiotics 2003, 56, 492.
(3) For prostaglandin reviews, see: (a) Roche, S. P.; Aitken, D. J. Eur.
J. Org. Chem. 2010, 5339. (b) Das, S.; Chandrasekhar, S.; Yadav, J. S.;
Gree, R. Chem. Rev. 2007, 107, 3286.
Scheme 3. Mechanistic Hypothesis for Gold-Catalyzed
Conversion of Acyloxyalkynyloxiranes 1 into
Cyclopentenones 3
(4) Belsito, D.; Bickers, D.; Bruze, M.; Calow, P.; Dagli, M. L.;
Dekant, W.; Fryer, A. D.; Greim, H.; Miyachi, Y.; Saurat, J. H.; Sipes, I.
G. Food Chem. Toxicol. 2012, 50, 5517.
(5) (a) Shimada, N.; Stewart, C.; Tius, M. A. Tetrahedron 2011, 67,
5851. (b) Grant, T. N.; Reider, C. J.; West, F. G. Chem. Commun.
2009, 5676. (c) Nakanishi, W.; West, F. G. Curr. Opin. Drug Discovery
Dev. 2009, 12, 732. (d) Frontier, A. J.; Collison, C. Tetrahedron 2005,
61, 7577. (e) Tius, M. Eur. J. Org. Chem. 2005, 2193.
(6) For selected examples, see: (a) Kerr, D. J.; Flynn, B. L. Org. Lett.
2012, 14, 1740. (b) Malona, J. A.; Cariou, K.; Frontier, A. J. J. Am.
Chem. Soc. 2009, 131, 7560. (c) He, W.; Huang, J.; Sun, X.; Frontier,
A. J. J. Am. Chem. Soc. 2007, 129, 498. (d) Williams, D. R.; Robinson,
L. A.; Nevill, C. R.; Reddy, J. P. Angew. Chem., Int. Ed. 2007, 46, 915.
(7) For recent reviews on Nazarov-type reactions, see: (a) Vaidya, T.;
Eisenberg, R.; Frontier, A. J. ChemCatChem 2011, 3, 1531. (b) Spencer,
W. T., III; Vaidya, T.; Frontier, A. J. Eur. J. Org. Chem. 2013, 3621.
(8) For gold-catalyzed Nazarov-type reactions, see: (a) Zhang, L.;
Wang, S. J. Am. Chem. Soc. 2006, 128, 1442. (b) Shi, F. Q.; Li, X.; Xia,
Y.; Zhang, L.; Yu, Z. X. J. Am. Chem. Soc. 2007, 129, 15503.
intramolecular [1,4]-addition of the acyloxy function via
oxophilic or carbophilic Au activations of acyloxyalkynyloxir-
anes could lead to the formation of the gold allenolate A, which
is in equilibrium with carbocation form B.15 Both could be
trapped by a 4-exo-dig cyclization, leading to the oxete C.16
Such strained intermediate should rapidly evolved into the
divinyl ketone 2 through cycloreversion. The gold-assisted
Nazarov reaction favored by electron-donating oxygen at the α-
carbon5d could next produce the oxocarbenium D. The latter
could evolve by intramolecular migration of the acyl part
furnishing cyclopentenone 3. The hydroxycyclopentenone 4
could be produced by hydrolysis of intermediate D, although in
situ deprotection of 3 cannot be excluded.
In conclusion, we have reported for the first time that Au(I)
or (III) complexes were prone to efficiently catalyze a Nazarov
reaction from activated divinyl ketones. Moreover, we also
developed a new cascade reaction starting from acyloxyalkyny-
loxiranes, precursors of divinyl ketones, giving cyclopentenones
with excellent yields using an NHC gold(III) triflimide
complex. Further works on synthetic applications and an
asymmetric version of this Nazarov reaction are in progress in
our group.
(c) Lemier
̀
e, G.; Gandon, V.; Cariou, K.; Fukuyama, T.; Dhimane, A.-
L.; Fensterbank, L.; Malacria, M. Org. Lett. 2007, 9, 2207. (d) Lemier
̀
e,
G.; Gandon, V.; Cariou, K.; Hours, A.; Fukuyama, T.; Dhimane, A.-L.;
Fensterbank, L.; Malacria, M. J. Am. Chem. Soc. 2009, 131, 2993.
(9) (a) Krafft, M. E.; Vidhani, D. V.; Cran, J. W.; Manoharan, M.
Chem. Commun. 2011, 6707. (b) Jin, T.; Yamamoto, Y. Org. Lett. 2008,
10, 3137.
(10) Cordonnier, M.-C.; Blanc, A.; Pale, P. Org. Lett. 2008, 10, 1569.
(11) Jacques, B.; Kirsch, J.; de Frem
́
ont, P.; Braunstein, P.
Organometallics 2012, 31, 4654.
(12) For selected examples, see: (a) Brady, S. F.; Bondi, S. M.;
Clardy, J. J. Am. Chem. Soc. 2001, 123, 9900. (b) Wang, X.-J.; Liu, Y.-
B.; Li, L.; Yu, S.-S.; Lv, H.-N.; Ma, S.-G.; Bao, X.-Q.; Zhang, D.; Qu, J.;
Li, Y. Org. Lett. 2012, 14, 5688. (c) Hodgson, D. M.; Galano, J.-M.;
Christlieb, M. Tetrahedron 2003, 59, 9719.
(13) (a) Bee, C.; Leclerc, E.; Tius, M. A. Org. Lett. 2003, 5, 4927.
(b) Liang, G.; Gradl, S. N.; Trauner, D. Org. Lett. 2003, 5, 4931.
(c) Hutson, G. E.; Turkmen, Y. E.; Rawal, V. H. J. Am. Chem. Soc.
̈
2013, 135, 4988.
(14) Mezailles, N.; Ricard, L.; Gagosz, F. Org. Lett. 2005, 7, 4133.
(15) For gold-catalyzed reactions implying such type of inter-
mediates, see: (a) Hoffmann, M.; Miaskiewicz, S.; Weibel, J.-M.; Pale,
P.; Blanc, A. Beilstein J. Org. Chem. 2013, 9, 1774. (b) Kern, N.; Blanc,
A.; Miaskiewicz, S.; Robinette, M.; Weibel, J.-M.; Pale, P. J. Org. Chem.
2012, 77, 4323. (c) Blanc, A.; Alix, A.; Weibel, J.-M.; Pale, P. Eur. J.
Org. Chem. 2010, 1644.
ASSOCIATED CONTENT
* Supporting Information
Complete experimental procedures, characterization data, and
spectral data. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
(16) Oxete may not be the sole intermediate; see: Per
Lopez, C. S.; Marco-Contelles, J.; Faza, O. N.; Soriano, E.; de Lera, A.
R. J. Org. Chem. 2009, 74, 2982.
́
ez, A. G.;
AUTHOR INFORMATION
Corresponding Authors
́
■
*E-mail: ppale@unistra.fr.
*E-mail: ablanc@unistra.fr.
Notes
The authors declare no competing financial interest.
911
dx.doi.org/10.1021/ol403663j | Org. Lett. 2014, 16, 908−911