Gold(I)-catalyzed cycloisomerization of β-alkynylpropiolactones to substituted α-pyrones

Article abstract of DOI:10.1021/ol1023739

Substituted α-pyrones were straightforwardly synthesized in good to excellent yields by a new gold(I)-catalyzed rearrangement of β-alkynylpropiolactones.

Full text of DOI:10.1021/ol1023739

ORGANIC
LETTERS
2010
Vol. 12, No. 23
5362-5365
Gold(I)-Catalyzed Cycloisomerization of
ꢀ-Alkynylpropiolactones to Substituted
r-Pyrones
Thomas Dombray, Aure´lien Blanc,* Jean-Marc Weibel, and Patrick Pale*
Laboratoire de Synthe`se et Re´actiVite´ Organiques Associe´ au CNRS, Institut de
Chimie, UniVersite´ de Strasbourg, 4 rue Blaise Pascal, 67000 Strasbourg, France
ablanc@unistra.fr; ppale@unistra.fr
Received July 3, 2010
ABSTRACT
Substituted r-pyrones were straightforwardly synthesized in good to excellent yields by a new gold(I)-catalyzed rearrangement of
ꢀ-alkynylpropiolactones.
Late transition metals, especially silver and gold, proved very
efficient catalysts for various organic transformations over
the past decade.1,2 Among the latter, cycloisomerizations hold
a special position, often offering a rapid entry to function-
alized heterocycles. The cycloisomerization of alkynylox-
iranes is a typical example, providing a powerful method to
generate substituted furans. Silver3 and gold,4 but also
platinum5 and mercury,6 have been successfully employed
as catalysts for this transformation, although different mech-
anisms have been proposed (Scheme 1, top).
Scheme 1. Hypothesis for Transition-Metal-Catalyzed
Conversion of ꢀ-Alkynylpropiolactones to R-Pyrones
(1) For recent reviews on silver chemistry, see: (a) Weibel, J.-M.; Blanc,
A.; Pale, P. Chem. ReV. 2008, 108, 3149–3173. (b) Alvarez-Corral, M.;
Munoz-Dorado, M.; Rodriguez-Garcia, I. Chem. ReV. 2008, 108, 3174–
3198. (c) Yamamoto, Y. Chem. ReV. 2008, 108, 3199–3222. (d) Patil, N. T.;
Yamamoto, Y. Chem. ReV. 2008, 108, 3395–3442. (e) Halbes-Letinois, U.;
Weibel, J.-M.; Pale, P. Chem. Soc. ReV. 2007, 36, 759–769
.
(2) For selected recent reviews on gold chemistry, see: (a) Li, Z.;
Brouwer, C.; He, C. Chem. ReV. 2008, 108, 3239–3265. (b) Arcadi, A.
Chem. ReV. 2008, 108, 3266–3325. (c) Jime´nez-Nunez, E.; Echavarren,
A. M. Chem. ReV. 2008, 108, 3326–3350. (d) Gorin, D. J.; Sherry, B. D.;
Toste, F. D. Chem. ReV. 2008, 108, 3351–3378. (e) Hashmi, A. S. K.;
Rudolph, M. Chem. Soc. ReV. 2008, 37, 1766–1775. (f) Hashmi, A. S. K.
Chem. ReV. 2007, 107, 3180–3211. (g) Jime´nez-Nu´n˜ez, E.; Echavarren,
A. M. Chem. Commun. 2007, 333–346. (h) Hashmi, A. S. K.; Hutchings,
On the basis of our precedents in such reactions,3,7 we
reasoned that incrementation of the epoxide ring in alkynyl-
oxiranes, i.e., alkynyloxetanes, would lead to the correspond-
ing homologated heterocycle, i.e., pyrans. With less strain,
alkynyloxetanes would be less prone to ring opening before
G. J. Angew. Chem., Int. Ed. 2006, 45, 7896–7936
.
(3) Blanc, A.; Tenbrink, K.; Weibel, J.-M.; Pale, P. J. Org. Chem. 2009,
74, 4360–4363.
(4) Hashmi, A. S. K.; Sinha, P. AdV. Synth. Catal. 2004, 346, 432–438.
(5) (a) Yoshida, M.; Al-Amin, M.; Matsuda, K. S. Tetrahedron Lett.
2008, 49, 5021–5023. (b) Yoshida, M.; Al-Amin, M.; Shishido, K. Synthesis
2009, 2454–2466.
(7) (a) Blanc, A.; Tenbrink, K.; Weibel, J.-M.; Pale, P. J. Org. Chem.
2009, 74, 5342–5348. (b) Blanc, A.; Alix, A.; Weibel, J.-M.; Pale, P. Eur.
J. Org. Chem. 2010, 1644–1647.
(6) Miller, D. J. Chem. Soc. C 1969, 12–15.
10.1021/ol1023739  2010 American Chemical Society
Published on Web 11/11/2010

cyclization as shown for alkynyloxiranes,7 and thus, other
or new mechanisms would occur, improving our mechanistic
knowledge of Au-catalyzed reactions.8 For further synthetic
purpose, we chose ꢀ-alkynylpropionolactones (4-alkynylox-
etan-2-ones) as starting materials. Such compounds should
give R-pyrones upon Ag- or Au-catalyzed cycloisomerization
(Scheme 1, bottom).
Table 1. Screening of Catalysts for the Transformation of
ꢀ-Alkynylpropiolactones 1a
R-Pyrones (2-pyrones or pyran-2-ones) indeed are an inter-
esting class of compounds in organic chemistry as such
structures are found in many natural products and in various
pharmacologically active compounds.9 They exhibit a wide
range of biological activities. Therefore, any new synthetic
method affording highly substituted R-pyrones under mild
conditions would be valuable in this context, especially since
the reported procedures usually required several steps with a
relatively modest efficiency (overall yields often lower than
40%).9
time
(h)
yield
(%)b
entry catalyst (5 mol %)
conditionsa
CH2Cl2, rfx
1
AgOTf
48
8
2
3
4
AgOTf/pTsOH
PPh3AuXd
AuCl
CH2Cl2, rfx
24/48 tracesc
e
CH2Cl2/MeOH, rt 48
-
1c f
,
CH2Cl2, rfx
CH2Cl2, rfx
CH2Cl2, rfx
CH2Cl2, rfx
CH2Cl2, rfx
CH2Cl2, rfx
CH2Cl2, rt
CH2Cl2, rt
CH3CN, rfx
Toluene, rfx
48
48
48
3
1
3
5
5
16
1
5
,
5
AuCl3
4c f
6
7
8
9
10
11
12
13
14
15
16
17
PPh3AuCl
PPh3AuBF4
PPh3AuSbF6
PPh3AuOTf
PPh3AuSbF6
PPh3AuOTf
PPh3AuOTf
PPh3AuOTf
(pCF3Ph)3PAuOTf CH2Cl2, rt
(pCF3Ph)3PAuOTf CH2Cl2, rfx
(F5Ph)3PAuOTf
(F5Ph)3PAuOTf
tracese
53
We report here that gold(I) catalyzes the rearrangement
of ꢀ-alkynylpropiolactones to substituted R-pyrones, provid-
ing an unprecedented route to such heterocycles10 in a two-
step and high overall yield procedure.
75
81
60
54
g
-
ꢀ-Lactones11 can be conveniently prepared through the
cyclocondensation of acyl halides and propargyl aldehydes
using an achiral procedure derived from the reaction
described by Nelson and co-workers.12 Despite the inherent
fragility of these substrates, isolated yields were routinely
higher than 60% (Scheme 2).
74
65
81
71
74
3
2.5
1
CH2Cl2, rt
CH2Cl2, rfx
a Reactions run under argon, C ) 0.1 mol/L. b Yields of pure isolated
products, unless otherwise stated. c The starting material was recovered.
d Various anions have been checked (X ) Cl, SbF6, OTf, NTf2). e Degradation
occurred. f Yields of 2a were estimated from crude 1H NMR spectra.
g Decarboxylation occurred (see text).
Scheme 2. Direct Synthesis of ꢀ-Alkynylpropiolactones
Applying our best conditions for the Ag- or Au-catalyzed
alkynyloxirane cycloisomerization in the presence or not of
a proton source3 or nucleophile7 mostly returned the starting
materials or led to decomposition (entries 1-3). Neverthe-
less, the expected pyrone 2a could be isolated from Ag-
catalyzed reactions, although in very low yield, even in
refluxing dichloromethane (entry 1). Switching to simple gold
chloride catalysts in pure dichloromethane did not help and
led to almost no transformation despite the detection of the
expected pyrone (entries 4 and 5). Although more soluble,
the corresponding triphenylphosphinogold chloride was not
reactive, leaving mostly untouched the starting lactone 1a,
even after prolonged heating (entry 6). In contrast, cationic
gold complexes in situ prepared from the chloride complex
gave the expected R-pyrones in variable yields depending
on the counteranion (entries 7-9). As expected, higher
temperatures sped up the reaction and often increased yields
(e.g., entries 8 vs 10 and 9 vs 11). Heat could also favor the
formation of side products, ꢀ-lactones usually leading to
enynes resulting from retro-[2 + 2]-cycloaddition (4 in
Scheme 3).11 However, the main side product formed in the
presence of gold was the conjugated acid 3a and surprisingly
not the retro-[2 + 2]-cycloaddition enyne product 4a
(Scheme 3). The best compromise between rearrangement
To establish appropriate reaction conditions, we subjected
the 3-(1-heptynyl)propiolactone 1a to various gold catalysts
expecting the formation of the 6-pentyl-2-pyrone 2a (Table
1). It is noteworthy that pyrone 2a is a natural product, first
identified from Trichoderma Viride,9,13 exhibiting a charac-
teristic coconut-like aroma as well as antibiotic and antifungal
properties.14
(8) For the sole review on Au-catalyzed reaction mechanisms, see:
Hashmi, A. S. K. Angew. Chem., Int. Ed. 2010, 49, 5232–5241.
(9) For reviews on R-pyrones, see: (a) Goel, A.; Ram, V. J. Tetrahedron
2009, 65, 7865–7913. (b) McGlacken, G. P.; Fairlamb, I. J. S. Nat. Prod.
Rep. 2005, 22, 369–385.
(10) A synthesis of R-pyrones has recently been reported, based on gold-
catalyzed [3 + 3]-sigmatropic rearrangement and cyclization. See: (a) Luo,
T.; Schreiber, S. L. Angew. Chem., Int. Ed. 2007, 46, 8250–8253. (b) Luo,
T.; Schreiber, S. L. J. Am. Chem. Soc. 2009, 131, 5667–5674.
(11) For reviews on ꢀ-lactones, see: (a) Pommier, A.; Pons, J.-M. Synthesis
1993, 441–459. (b) Yang, H. W.; Romo, D. Tetrahedron 1999, 55, 6403–
6434. (c) Schneider, C. Angew. Chem., Int. Ed. 2002, 42, 744–746.
(12) See Supporting Information: (a) Nelson, S. G.; Peelen, T. J.; Wan,
Z. J. Am. Chem. Soc. 1999, 121, 9742–9743. (b) Nelson, S. G.; Wan, Z.
Org. Lett. 2000, 2, 1883–1886.
(14) (a) Simon, A.; Dunlop, R. W.; Ghisalberti, E. L.; Sivasithamparam,
K. Soil Biol. Chem. 1988, 20, 263–264. (b) Scarselletti, R.; Faull, J. L.
Mycol. Res. 1994, 98, 1207–1209.
(13) Collins, R. P.; Halim, A. F. J. Agric. Food Chem. 1972, 20, 437–
438
.
Org. Lett., Vol. 12, No. 23, 2010
5363

Scheme 3
.
Competitive Reactions Sometimes Observed during
Table 2. Scope of Gold(I)-Catalyzed R-Pyrone Formation
the ꢀ-Alkynylpropiolactone to Pyrone Rearrangement
efficiency and byproduct formation was achieved with
triphenylphosphinogold triflate in refluxing dichloromethane
(entry 9). With this catalyst, more coordinating solvents
proved deleterious, inhibiting the reaction (entry 12), while
in apolar noncoordinating solvent, the reaction went faster
but some decarboxylation also occurred probably due to the
higher temperature (entry 13). It is worth noting that
decarboxylation was the only process in refluxing acetonitrile
(entry 12).
More electrophilic complexes gave the same trends. With
the tris(4-trifluoromethylphenyl)phosphine as the gold ligand,
results almost identical to those achieved with triphenylphos-
phinogold triflate were obtained (entries 15 vs 9) but only
in refluxing dichloromethane (entries 14 vs 15). The more
electron-withdrawing tris(pentafluorophenyl)phosphine ren-
dered the catalyst more reactive and thus induced a faster
reaction (entries 16 and 17). As before, heat slightly
improved the reaction efficiency, but again side reactions
became competitive (entries 17 vs 16).
Control experiments showed the stability of the 2-pyrone
2a under the conditions used, confirming that the side
products, formed under certain conditions, originated from
the starting ꢀ-lactone 1a.
Since two sets of conditions emerged from this screening
(entries 9 and 14), we submitted various alkynyl-ꢀ-lactone
derivatives to both conditions, to examine the scope of this
new Au-catalyzed rearrangement (Table 2).
Simple alkyl chains and cycloalkyl derivatives 1a-e gave
the expected 6-substituted 2-pyrones in good to high yields
(entries 1-10). With a substituent adjacent to the ꢀ-lactone
carbonyl group, the pyrone formation was surprisingly slow
(entries 3, 4 vs 1, 2 and 9, 10 vs 7, 8), and the larger the
group, the larger this slowing effect (entries 5, 6 vs 3, 4).
Aryl groups tended to destabilize the starting ꢀ-lactones
as well as the corresponding 2-pyrones, and thus these
reactions led to more side products, lowering yields (entries
11-18 vs 1-10), even at room temperature or at 0 °C
(entries 11-14). Both the corresponding enynes and the
conjugated acids (4 and 3 in Scheme 3) could be isolated or
detected (respectively, 1-8% and 2-10%) depending on the
conditions and the substrate.
a Method A: Ph3PAuOTf (5 mol %). Method B: (pCF3Ph)3PAuOTf (5
mol %). b Yields of pure isolated products. c Performed at 0 °C. d The
reaction performed at 0 °C gave a lower yield (46%). e No conversion.
and 8% for 3f with (pCF3Ph)3PAuOTf as the catalyst. With
a methyl adjacent to the carbonyl group, the reactions were
again slower, and heat improved yield when (pCF3Ph)3-
PAuOTf was used as catalyst (entries 14 vs 13). Interestingly,
adding an electron-withdrawing group on the phenyl ring
stabilized the ꢀ-lactone and restored its reactivity toward gold
(entries 15-16 vs 11-12). However, if this group was
strongly electron-withdrawing, no evolution occurred (e.g.,
p-nitrophenyl; not mentioned in Table 2). Such behavior is
reminiscent from the electronic effects observed in decar-
boxylation of ꢀ-lactones.15
The ꢀ-phenylethynyl-ꢀ-lactone 1f gave the expected
6-phenyl-2-pyrone 2f (entries 11 and 12) together with
1-phenylbut-3-en-1-yne 4f and the conjugated acid 3f. With
Ph3PAuOTf as catalysts, the latter could be isolated in,
respectively, 8 and 10% yield, while only as traces for 4f
5364
Org. Lett., Vol. 12, No. 23, 2010

Ether substituents at the propargylic position dramatically
slowed the reaction (entries 19-20 vs 1-2) and favored the
side formation of the unsaturated acid 3 (28% of 3j). The
concomitant presence of a substituent R to the ꢀ-lactone
carbonyl group slowed even more the reaction, so that almost
no evolution occurred (entries 21-22 vs 19-20). Unexpect-
edly, a decrease in reaction time was also observed if the
same ether substituent was remote from the propargylic
position (entries 23-24 vs 19-20 vs 1-4). When this benzyl
ether was replaced by a pivaloyl ester, the normal behavior
was restored, and the corresponding pyrone was obtained in
high yield within a few hours (entries 25-26 vs 23-24).
The same normal behavior was also observed when this
benzyl ether was far remote from the alkyne moiety (entries
27-28). These results revealed a special role of the benzyl
ether unit when placed relatively close from the Au-
coordinating sites, reminiscent of Au-catalyzed Friedel-Crafts
reactions,16 or from the benzyloxy fragmentation recently
reported.17,18
It is worth noting that most derivatives with a substituent
adjacent to the ꢀ-lactone carbonyl group proved far less
reactive with a concomitant increase in reaction time (entries
3-4 vs 1-2, 5-6 vs 1-2, 9-10 vs 7-8, 13-14 vs 11-12,
17-18 vs 15-16, and 21-22 vs 19-20). Moreover, large
groups reduced the reactivity more strongly than small groups
(entries 5-6 vs 3-4 vs 1-2). These series of disubstituted
ꢀ-lactones were produced as an approximate 3:1 mixture of
diastereoisomers, but no significant difference was observed
in their evolution, whatever the catalyst used. The results
obtained with this series suggested some hindrance at some
stage of the rearrangement, most probably at the coordination
step (vide infra).
From a mechanistic point of view, it is not possible to
naively extend the mechanism we proposed for alkynyl
epoxide rearrangement after our detailed investigations.7a In
the latter, a nucleophile, either introduced in purpose or
adventitiously present, opened the epoxide, and the resulting
alkoxide then cyclized in a gold-catalyzed process. In the
present case, running the alkynyl-ꢀ-lactone to R-pyrone
rearrangement in the presence of a nucleophile proved
deleterious, while without it, the reaction proceeded well (see
Table 1). Nucleophilic opening of alkynyl-ꢀ-lactone and Au-
catalyzed cyclization is thus not the major pathway.
The general trend observed with an extra group close to
the ꢀ-lactone carbonyl group on one hand and the role of
some alkynyl substituent on the other hand suggested changes
in coordination ability and thus supported equilibrium
between σ- and π-Au complexes (Scheme 4, A and B).19
Scheme 4. Proposed Mechanism for the Au(I)-Catalyzed
Rearrangement of ꢀ-Alkynylpropiolactones to R-Pyrones
The formation of side products, especially the acid 3 (see
Scheme 3), also supported σ-coordination at the ꢀ-lactone
carbonyl. The resulting σ-Au ꢀ-lactone complexes B could
be more prone to decarboxylation, especially when fully
conjugated enynesse.g., 4f,g (R′ ) Ph)scould be obtained
(Table 2, entries 11-18). Such σ-Au complexes B could also
evolve to an open cationic gold carboxylate C, a precursor
of the enynoic acids 3 (Scheme 4). It is worth mentioning
that such cationic intermediates have been postulated to
rationalize the stereochemical outcome of ꢀ-lactone decar-
boxylation20 and must be involved in the Lewis acid
promoted ring expansion of ꢀ-lactones.21 Both σ- and π-Au
complexes A and B would lead to cationic pyrone gold
intermediate D, possibly through either a 1,3-oxygen shift
or Hashmi-type cyclization4 from A and through cyclization
of C from B. The intermediate D would then give the
corresponding pyrones 2 upon proton elimination and
subsequent protodeauration (Scheme 4).22
In conclusion, we have reported a novel synthesis of
R-pyrones through Au-catalyzed cycloisomerization of ꢀ-alky-
nylpropiolactones. Since the latter can be obtained by
condensing acyl chlorides to aldehydes, this synthesis is
particularly appealing, providing in two steps with good to
high yields a wide diversity of R-pyrones starting from very
simple compounds.
Acknowledgment. The authors thank the CNRS and the
French Ministry of Research for financial support. T.D.
thanks the French Ministry of Research for a PhD fellowship.
Supporting Information Available: Selected experimen-
tal procedures and spectral data. This material is available
free of charge via the Internet at http://pubs.acs.org.
(15) Krabbenhoft, H. O. J. Org. Chem. 1978, 43, 1305–1307.
(16) For some recent examples, see: (a) Bandini, M.; Eichholzer, A.
Angew. Chem., Int. Ed. 2009, 48, 9533–9537. (b) Lu, Y.; Du, X.; Jia, X.;
Liu, Y. AdV. Synth. Catal. 2009, 351, 1517–1522. (c) Zhang, Z.; Shi, M.
Chem.sEur. J. 2010, 16, 7725–7729.
OL1023739
(20) (a) Mulzer, J.; Zippel, M.; Bru¨ntrup, G. Angew. Chem., Int. Ed.
1980, 19, 465–467. (b) Moyano, A.; Pericas, M. A.; Valenti, E. J. Org.
Chem. 1989, 54, 573–581. (c) Asencio, G.; Miranda, M. A.; Perez-Prieto,
J.; Sabater, M. J.; Simon-Fuentes, A. Angew. Chem., Int. Ed. 1990, 29,
1146–1148.
(17) Bolte, B.; Odabachian, Y.; Gagosz, F. J. Am. Chem. Soc. 2010,
132, 7294–7296
.
(18) It is worth noting that no benzaldehyde17 could be detected in the
crude mixture by NMR, but modified aromatic compounds were detected
(e10%).16
(21) (a) Mulzer, J.; Bru¨ntrup, G. Angew. Chem., Int. Ed. 1979, 18, 793–
795. (b) Black, T. H.; Hall, J. A.; Sheu, R. G. J. Org. Chem. 1988, 53,
2371–2378.
(19) (a) Yamamoto, Y. J. Org. Chem. 2007, 72, 7817–7831. (b) Pe´rez,
A. G.; Lo´pez, C. S.; Marco-Contelles, J.; Faza, O. N.; Soriano, E.; de Lera,
A. R. J. Org. Chem. 2009, 74, 2982–2991.
(22) Krauter, C. M.; Hashmi, A. S. K.; Pernpointner, M. Chem. Catal.
Chem. 2010, 2, 1226–1230.
Org. Lett., Vol. 12, No. 23, 2010
5365