Gold-catalyzed meyer-schuster rearrangement: Application to the synthesis of prostaglandins

Article abstract of DOI:10.1021/om1005534

Synthetic pathway paved with gold: An improved protocol for the Meyer-Schuster rearrangement is presented using novel gold catalysts. This method is used to provide a straightforward synthetic approach to prostaglandins.

Full text of DOI:10.1021/om1005534

Organometallics 2010, 29, 3665–3668 3665
DOI: 10.1021/om1005534
Gold-Catalyzed Meyer-Schuster Rearrangement: Application to the
Synthesis of Prostaglandins
†
Ruben S. Ramon, Sylvain Gaillard, Alexandra M. Z. Slawin, Alessio Porta,
†
†
‡
ꢀ
ꢀ
Alessandro D’Alfonso,‡ Giuseppe Zanoni,‡ and Steven P. Nolan*,†
†EaStCHEM School of Chemistry, University of St Andrews, St Andrews, KY16 9ST, U.K., and
‡Department of Organic Chemistry, University of Pavia, Viale Taramelli, 10 27100 Pavia, Italy
Received June 4, 2010
The Meyer-Schuster (M-S) rearrangement has recently
been shown to be highly efficient in the formation of
conjugated enones.1 In the course of developing suitable
gold-based catalysts for this reaction, the application
of [(IPr)AuCl]/AgSbF6 (IPr = N,N0-bis(2,6-diisopropyl-
phenyl)imidazol-2-ylidene) was reported as a useful catalytic
system.2 In the context of the mechanism of the M-S reaction
mediated by gold-NHC complexes, we proposed the initial
formation of [(IPr)Au(OH)] (1) as the active catalytic spe-
cies.2 To support this mechanistic hypothesis, [(IPr)Au(OH)]
(1) was targeted for synthesis and recently isolated and fully
characterized.3 Surprisingly, the catalytic activity of 1 in the
M-S rearrangement was found to be modest. Further syn-
thetic studies on this system permitted the isolation of a novel
dinuclear complex, [{(IPr)Au}2(μ-OH)]BF4 (2), in aqueous
media.4 It appeared logical to consider 2 as an active species
in the M-S rearrangement, as the catalytic reaction is per-
formed in a water-containing medium. Herein, the activity of
digold complexes in the M-S rearrangement, performed at
room temperature and affording high turnover numbers
(TONs), is reported and linked to the gold cationic species
formed when 1 is reacted with an excess of acid. Addition-
ally, the superior catalytic behavior of 2 in the M-S rear-
rangement permits its use in the development of a mole-
cular assembly strategy leading to the efficient synthesis of
prostaglandins.
In order to optimize the M-S rearrangement using gold
catalysts, reactivity studies were performed using the well-
defined [(IPr)Au(OH)] (1) complex and a “protic acid acti-
vator”. This strategy is presently being examined for numer-
ous cationic gold(I)-mediated transformations.4 As shown in
Table 1, 1 alone led to very low conversion even when pro-
longed reaction times and higher temperatures were used
(Table 1, entries 1 and 2). This clearly excludes 1 as a reactive
species in the M-S reaction. On the other hand, addition of
HBF4 to 1 results in a rapid conversion to product. In this
instance, the in situ formation of [{(IPr)Au}2(μ-OH)]BF4 (2)
(Table 1, entry 3) or [(IPr)Au]BF4 (Table 1, entry 5) is observed,
and one or the other of the two species is formed as a function
of the amount of acid used.4
As [{(IPr)Au}2(μ-OH)]BF4 (2) is easily formed in aqueous
media, 2 was foreseen as a potentially highly active catalyst
for the M-S rearrangement, although a comparison of entries
3 and 5 in Table 1 indicates that slightly longer reaction times
are required for 2 than for [Au(IPr)]BF4. As hoped, the well-
defined 2 (Table 1, entry 8) allowed for very good yields of
product in short reactions times compared to the use of
[(IPr)Au(NTf2)] (3) (Table 1, entry 10).
Complex 3 has already proven to be as an excellent and
stable catalyst due to its inner-sphere NTf2 (NTf2 = bis-
(trifluoromethanesulfonyl)imidate) ligand and the apparent
ease of this ligand to leave the coordination sphere of gold-
(I) to generate a vacant site.5 Using the synthetic strategy
outlined above for the synthesis of 2, the synthesis of [{(IPr)-
Au}2(μ-OH)]NTf2 (4) was attempted to validate the general-
ity of our synthetic protocol but also to possibly shed light on
the fate of 3 in the course of catalytic reactions. Additionally,
the isolation of such a dinuclear species would support a
developing mechanistic hypothesis suggesting the initial
formation of hydroxy-bridged digold complexes in catalysis
and would present a rare example where NTf2 is an outer-sphere
ligand.6 [{(IPr)Au}2(μ-OH)]NTf2 (4) was indeed synthesized
in an analogous manner to that of 2, where 1 was added under
nitrogen to a suspension of 0.5 equiv of HNTf2 in anhydrous
toluene at room temperature. After appropriate workup,
4 was isolated as a microcrystalline white solid in quantitative
yield. The structure of 4 was confirmed by single X-ray
diffraction study (Figure 1).7
To test the scope of the M-S methodology with the most
interesting catalyst/precatalysts on hand, the reactivity of
1 (with 1.5 equiv of HBF4), 2, [(IPr)Au(NTf2)] (3), and
[{(IPr)Au}2(μ-OH)]NTf2 (4) was tested for a small series of
characteristic propargylic alcohols (Table 2). Catalysts 2-4
have the advantage of avoiding the use of the air-, light-, and
moisture-sensitive silver salts of type AgX (X = BF4, SbF6,
OTf, PF6) to abstract chloride from the usually employed
[(L)AuCl] (L = ligand) complexes.8,9 The use of well-defined
complexes or of simple acid activation results in a simplified
* To whom correspondence should be addressed. Fax: þ44(0)1334
463808. E-mail: snolan@st-andrews.ac.uk.
(1) For a recent review on the Meyer-Schuster rearrangement, see:
Cadierno, V.; Crochet, P.; Garcı
´
a-Garrido, S. E.; Gimeno, J. Dalton
(5) Ricard, L.; Gagosz, F. Organometallics 2007, 26, 4704–4707.
(6) For one example of an outer-sphere NTf2 ligand in gold com-
ꢀ
plexes, see: Weber, D.; Tarselli, M. A.; Gagne, M. R. Angew. Chem., Int.
Trans. 2010, 39, 4015-4031. See also references cited therein.
ꢀ
(2) Ramon, R. S.; Marion, N.; Nolan, S. P. Tetrahedron 2009, 65,
1767–1773.
Ed. 2009, 48, 5733–5736.
(3) Gaillard, S.; Slawin, A. M. Z.; Nolan, S. P. Chem. Commun. 2010,
46, 2742–2744.
ꢀ
(4) Gaillard, S.; Bosson, J.; Ramon, R. S.; Nun, P.; Nolan, S. P.
Manuscript submitted.
(7) CCDC 772704 (4) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallograophic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
r
2010 American Chemical Society
Published on Web 07/15/2010
pubs.acs.org/Organometallics

ꢀ
Ramon et al.
3666 Organometallics, Vol. 29, No. 16, 2010
Table 1. Influence of HBF4 on the Catalytic Activity of 1 and 2
Table 2. Screening of Catalysts with Representative
Propargylic Alcohols
entry
catalysta
acid (mol %)
time (h)
conversion (%)b
1
2
3
4
5
6
7
8
9
10
1
1
1
1
0
0
0.5
48
0.5
0.5
0.5
1
0.5
1
0.5
3
9
29c
73
98
99
0
HBF4 (1)
HBF4 (2)
HBF4 (3)
HBF4 (20)
0
1
none
2
2
2
3
61
99
99
99
0
HBF4 (1)
0
a Catalysts: [(IPr)Au(OH)] (1), [{(IPr)Au}2(μ-OH)]BF4 (2). b Only the
E-isomer was observed. c 60 °C.
a Catalysts: 1 [(IPr)Au(OH)], 2 [{(IPr)Au}2(μ-OH)]BF4, 3 [(IPr)Au-
(NTf2)], 4 [{(IPr)Au}2(μ-OH)]NTf2. b Isolated yields; reaction was moni-
tored by TLC until no further conversion was observed. c 1.5 equiv
referring to Au. d The allenol tautomer was isolated as a side product:
26% (2), 6% (3), 21% (4). e nd = not determined. The target molecule
was obtained in an inseparable mixture of products.
catalytic activity than other catalysts examined. It should be
noted that catalysts 2 and 3 lead in the case of 1-cyclohex-
enylhept-1-yn-3-ol to an inseparable mixture of the desired
M-S product and a second product resulting from the re-
action with methanol (Table 2, entries 9, 11). Changing the
solvent to DCM or THF resulted in a complete loss of re-
activity. In order to address the issues of catalyst stability and
decomposition, the behavior of [{(IPr)Au}2(μ-OH)]BF4 (2)
at low catalyst loadings was examined. To probe these im-
portant considerations, a reaction was conducted with se-
quential addition of starting material. Remarkably in this
series of experiments, 2 proved extremely stable, resilient
to decomposition, and active under catalytic conditions
(0.03 mol % of 2, TON 3330) when the reaction was conti-
nuously fed substrate for 3 weeks!
Figure 1. X-ray structure of [{(IPr)Au}2(μ-OH)]NTf2, 4. Selec-
ted bond lengths [A] and angles [deg]: Au1-Au2 3.6348(12);
Au1-O1, 2.036(4); Au2-O1, 2.043(4); Au1-C1, 1.959(5); Au2-
C31, 1.954(5); Au1-O1-Au2, 126.0(2); O1-H1O, 0.988(19) ;
Au1-O1-H1O, 108(3); Au2-O1-H1O, 123(3); C1-Au1-O1,
175.26(19); C31-Au2-O1, 176.23(17).
To finally demonstrate the use of the gold-catalyzed
Meyer-Schuster rearrangement in synthesis, the catalytic
system was applied to an assembly strategy leading to the
straightforward synthesis of prostaglandins. Natural prosta-
glandins are unsaturated fatty acids derived from arachi-
donic acid. Many classes of prostaglandins are characterized
and more economically attractive catalytic protocol. First
results on some benchmark substrates revealed excellent re-
activity and high isolated yields using reaction conditions
previously reported,2 but here reactions are conducted at
room temperature (Table 2).
As shown in Table 2, the reactivity of all systems examined
leads to very similar results for the corresponding substrates.
Interestingly, [(IPr)Au(NTf2)] (3) exhibited slightly reduced
by very high biological activity.10 Prostaglandin F2R (PGF2R
,
Figure 2), for example, is known to be a very potent vaso-
constrictor and oxytoxic agent.11 Most of the synthetic pro-
cedures for prostaglandin synthesis employ one of the
(8) Of note, the use of a co-catalystperforming chloride abstraction in
gold catalysts is not strictly required for some catalytic systems: For an
example, see: Sromek, A. W.; Rubina, M.; Gevorgyan, V. J. Am. Chem.
Soc. 2005, 127, 10500–10501.
(9) For previous examples of gold catalyst acid activation, see: (a) Teles,
J. H.; Brode, S.; Chabanas, M. Angew. Chem., Int. Ed. 1998, 37, 1415–1418.
(b) Mizushima, E.; Sato, K.; Hayashi, T.; Tanaka, M. Angew. Chem., Int. Ed.
2002, 41, 4563–4565.
€
(10) (a) Hoffmann-Roder, A.; Krause, N. Angew. Chem., Int. Ed.
2004, 43, 1196–1216. (b) Miller, S. B. Semin. Arthritis Rheum. 2006, 36,
37–49.
(11) Smyth, E. M.; Burke, A.; FitzGerald, G. A. Goodman & Gilman’s
The Pharmacological Basis of Therapeutics, 11th ed.; Brunton, L. L., Ed.;
McGraw-Hill Medical Publishing Division, 2005; pp 653-667.

Note
Organometallics, Vol. 29, No. 16, 2010 3667
as a mixture of E:Z olefins in 17:1 ratio (de 88%), while the
present protocol delivered only the desired E-stereoisomer in
86% isolated yield! With 9 in hand, the synthesis proceeded
readily to the target prostaglandin PGF2R via classical prosta-
glandin chemistry15 in 80% isolated yield and as a single stereo-
isomer. Mechanistically, the excellent conversion of 6 without
additional water strongly suggests a concerted mecha-
nism. However, closer investigations with both monogold
complexes and digold complexes in anhydrous conditions
revealed significantly longer reaction times with 1-phenyl-
hept-2-yn-1-ol. As a consequence, the involvement of traces
of water in catalysis cannot be excluded. On the other hand,
experimental results clearly exclude an earlier proposal of the
in situ formation of 1 initiating the catalytic cycle. Instead, a
resting state of type 2 and 4 is, we feel, more relevant and may
simply act as a reservoir of the active [(IPr)Au]þ.16
In conclusion, an improved protocol for the Meyer-
Schuster rearrangement with dinuclear gold complexes al-
lowing for low catalyst loadings was presented. The new pro-
tocol has been successfully applied to the synthesis of prosta-
glandins. This study helps exclude earlier mechanistic
proposals and highlights the relevance and possibly more
significant role of digold complexes in gold catalysis.
Figure 2. Synthetic route to prostaglandins via gold-catalyzed
M-S rearrangement.
Experimental Section
Unless otherwise stated, manipulations were performed un-
der air. Solvents were of puriss. grade and used as received.
NMR spectra were recorded on 400 and 300 MHz spectrometers
at ambient temperature in CDCl3. Chemical shifts are given in
parts per million (ppm) with respect to TMS. 19F chemical shifts
are given in parts per million (ppm) with respect to CFCl3.
Synthesis of 2. [(IPr)Au(OH)] 1 (97 mg, 0.160 mmol) was
dissolved in benzene (2 mL), and tetrafluoroboric acid diethyl
ether complex (11.0 μL, 0.080 mmol) was added. The reaction
mixture was stirred 4 h at room temperature. Pentane was added
to the reaction to precipitate the product as a white solid. The
crude white product was recrystallized from CH2Cl2/pentane to
known variations of the Corey method, in which the side
chains are sequentially attached in a specific order to a
derivative of the commercially available (-)-Corey lactone
aldehyde 5.12
Thus, the Corey strategy involves the installation of the
lower side chain (ω-chain) via an archetypal non-atom-
economical13 approach: a Horner-Wadsworth-Emmons
(HWE) reaction of a suitable keto-phosphonate.
It was envisioned that gold catalysis, in the form of the
Meyer-Schuster rearrangement, could improve the atom-
economy of the ω-chain installation through a catalytic
rearrangement of propargylic alcohol 6 to key enone 7
(Figure 2). 6 was readily prepared from commercially avail-
able enantiopure 5 in 90% isolated yield. 6 was then treated
with a catalytic amount of (IPr)AuCl/AgSbF6 (2 mol %) as
reported in an earlier contribution.2 Within the optimization
process (see table in the SI), we discovered that catalyst 2 (2
mol %) afforded 7 (E-isomer) in 85% isolated yield, using
anhydrous DCM as a solvent, which was advantageous for
the following synthetic step. A standard mixture of MeOH/
water (10:1) produced 7 (E-isomer) in 86% yield with
catalyst 2 (2 mol %) and 84% with catalyst 4 (2 mol %),
respectively, while with the prior reported reaction con-
ditions2 lower yields (81%) were obtained. 7 was then ela-
borated via 8 to bis-TBS-protected allyl alcohol 9, involving
the asymmetric diisopinocamphenyl chloroborane (DIP-
Cl)-mediated ketone reduction14 followed by silylation of
the corresponding alcohol. The latter has been used by
Mulzer et al.15 for the asymmetric synthesis of PGF2R. It should
be pointed out that with the Mulzer approach 9 was obtained
1
give 92 mg (90%) of a white microcrystalline solid. H NMR
(400 MHz, CDCl3): δ 7.50 (t, J = 7.8 Hz, 4H), 7.26 (s, 4H), 7.24
(d, J = 7.8 Hz, 8H), 2.39 (sept, J = 6.9 Hz, 8H), 1.19 (d, J = 6.9
Hz, 24H), 1.11 (d, J = 6.9 Hz, 24H). 13C NMR (75 MHz,
CDCl3): δ 162.6, 145.4, 133.6, 130.7, 124.4, 124.2, 124.1, 28.6,
24.4, 23.8. 19F NMR (185 Hz): δ -154.90, -154.85. IR (cm-1):
3621, 3167, 3137, 3084, 2964, 2928, 2871, 1596, 1553, 1472, 1421,
1386, 1365, 1329, 1215, 1058, 947, 807, 762, 707, 581, 455. Anal.
Calcd: C 51.06 (50.87), H 5.27 (5.77), N 4.36 (4.39).
Synthesis of 4. Trifluoromethanesulfonimide (73 mg, 0.260
mmol) was dissolved under N2 in anhydrous toluene (2 mL), and
[(IPr)Au(OH)] 1 (313 mg, 0.519 mmol) was added. The reaction
mixture was stirred overnight at room temperature. Pentane
(10 mL) was added to the reaction to ensure complete precipita-
tion of the product. The white solid was collected by filtration,
washed with pentane (2 ꢀ 5 mL), and dried under vacuum to
give 381 mg (100%) of a white microcrystalline solid. 1H NMR
(300 MHz, CDCl3): δ 7.49 (t, J = 7.8 Hz, 4H), 7.24 (d, J = 7.8
Hz, 8H), 7.14 (s, 4H), 2.37 (spt, J = 6.8 Hz, 8H), 1.17 (d, J = 6.8
Hz, 24H), 1.11 (d, J = 6.8 Hz, 24H). 13C NMR (100 MHz,
CDCl3): δ 162.9, 145.6, 133.6, 133.7, 130.9, 124.3, 124.2, 120.1
(q, J = 321.7 Hz), 28.8, 24.6, 23.9. 19F NMR (282 Hz, CDCl3): δ
-79.2. Anal. Calcd: C 45.78 (45.81), H 4.92 (5.01), N 4.57 (4.77).
Meyer-Schuster Rearrangement. In a typical reaction, 1-phe-
nylhept-2-yn-1-ol (188 mg, 1 mmol) was charged in a vial, and
(12) For the total synthesis of PGF2R, see: Corey, E. J.; Noyori, R.
Tetrahedron Lett. 1970, 11, 311–313.
(13) (a) Trost, B. M. Science 1991, 254, 1471–1477. (b) Trost, B. M.
Angew. Chem., Int. Ed. Engl. 1995, 34, 259–281.
(14) Brown, H. C.; Ramachandran, P. V. Acc. Chem. Res. 1992, 25,
16–24.
(16) Dinuclear intermediates in gold catalysis have been proposed:
Cheong, P. H.-Y.; Morganelli, P.; Luzung, M. R.; Houk, K. N.; Toste,
F. D. J. Am. Chem. Soc. 2008, 130, 4517–4526.
(15) Sheddan, N. A.; Mulzer, J. Org. Lett. 2006, 8, 3101–3104.

ꢀ
Ramon et al.
3668 Organometallics, Vol. 29, No. 16, 2010
methanol (7.5 mL) was added. 2 (12.8 mg, 0.01 mmol) was added
followed by water (750 μL). The reaction was stirred at room
temperature until conversion reached completion. After eva-
poration of volatile compounds under reduced pressure, the
residue was purified via flash chromatography (pentane/Et2O,
95:5) to obtain 176 mg (93%) of a pale yellow solid. 1H NMR
(400 MHz, CDCl3): δ 7.60-7.50 (m, 3H), 7.42-7.36 (m, 3H),
6.75 (d, J = 16.2 Hz, 1H), 2.67 (t, J = 7.5 Hz, 2H), 1.74-1.60
(m, 2H), 1.48-1.31 (m, 2H), 0.95 (t, J = 7.3 Hz, 3H).
Acknowledgment. The ERC (FUNCAT-Advanced
Researcher grant to S.P.N.) and the EPSRC are gratefully
acknowledged for support of this work.
Supporting Information Available: Supporting Information
for this article including additional tables and characterization
of all compounds reported is available free of charge via the
Internet at http://pubs.acs.org.