Classical carbonyl reactivity enables a short synthesis of the core structure of acutumine

Article abstract of DOI:10.1016/j.tet.2007.03.024

The development of a direct synthesis of the complex core topology of the alkaloid acutumine from a simple keto proline derivative is described. An efficient sequence of three carbonyl-dependent reactions is at the heart of this design for synthesis.

Full text of DOI:10.1016/j.tet.2007.03.024

Tetrahedron 63 (2007) 6446–6453
Classical carbonyl reactivity enables a short synthesis
of the core structure of acutumine
*
Robert J. Moreau and Erik J. Sorensen
Frick Chemical Laboratory, Princeton University, Princeton, NJ 08544-1009, United States
Received 21 January 2007; revised 1 March 2007; accepted 2 March 2007
Available online 12 March 2007
Abstract—The development of a direct synthesis of the complex core topology of the alkaloid acutumine from a simple keto proline
derivative is described. An efficient sequence of three carbonyl-dependent reactions is at the heart of this design for synthesis.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
co-workers.5 The remarkable structure of acutumine (1)
(Fig. 1), with its complex cyclic connectivity and conspicu-
ous chlorine atom, consists of a propellane-like [4.3.3.0]
fused tricycle with appended spirocycle core and five con-
tiguous stereocenters, of which three are fully substituted.
In the solid state, the nitrogen atom and chlorine-bearing
The intrinsic reactivity of the carbonyl group and the special
properties it bestows on organic compounds have enabled
the development of many types of bond forming strategies
that rapidly produce molecular complexity and are of great
value to the field of chemical synthesis.1 In the context of
natural products that are both topologically and stereo-
chemically complex, our laboratory has been pursuing strat-
egies for synthesis that are founded on fundamental,
carbonyl-dependent reactivity. A synthesis of the tricyclo-
[4.3.3.0]dodecane core structure of acutumine featuring an
intramolecular rearrangement is the subject of this article.
˚
carbon of acutumine are separated by only 3.2 A.
H3CO
OH
Cl
O
N
CH3
OCH3
OCH3
O
The unique, chlorine-containing alkaloid acutumine is a
minor constituent of the plants Sinomenium acutum, Menis-
permum canadense, and Menispermum dauricum, the dry
rhizome of which (Rhizoma menispermi) is part of tradi-
tional Chinese medicine and is now officially included in
the Chinese Pharmacopoeia as an analgesic and fever reduc-
ing agent. In 2002, it was reported that acutumine selectively
inhibits human T-cell growth, although with low micromolar
potency.2 In 2004, a patent described that this alkaloid
has memory-enhancing properties in experimental animal
models.3
1: Acutumine
Figure 1. The molecular structure of acutumine (1).
In the wake of these studies, Barton and co-workers de-
scribed their idea that acutumine may originate from a simple
benzylisoquinoline alkaloid by a reaction sequence starting
with an oxidative coupling of two phenol radicals.6,7 Two
independent efforts to evaluate the experimental feasibility
of Barton’s creative hypothesis have appeared, and the
most recent of these investigations by Wipf and co-workers
has led to a slight modification of Barton’s proposal.8 The
recent, impressive construction of the propellane-like core
structure of acutumine by the Castle laboratory is the only
report describing progress toward a chemical synthesis of
this natural product.9
Acutumine was first isolated in a pure state in 1929 by Goto
and Sudzuki from the Kitasato Institute in Tokyo, Japan.4
Their early structural studies indicated that acutumine
contains a ketone, an N-methyl, and three O-methyl groups,
but nearly 40 years passed before the full structure and
absolute stereochemistry of this compound became known
through the X-ray crystallographic studies of Tomita and
2. Concepts for synthesis
Our early studies were founded on the idea that we might
achieve a synthesis of acutumine (1) from p-benzoquinone
(2) (Scheme 1). While it was possible to transform this abun-
dant starting material to enantiomerically enriched alcohol 3
Keywords: Acutumine; The carbonyl group; b-Elimination; Michael reac-
tion; Dieckmann condensation; Chlorine-containing alkaloids.
* Corresponding author. Tel.: +1 609 258 8135; fax: +1 609 258 1980;
e-mail: ejs@princeton.edu
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.024

R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
6447
by an efficient reaction sequence, all efforts to move the
latter compound closer to acutumine were unsuccessful.
With some hesitation, we abandoned the idea that our syn-
thesis should commence from p-benzoquinone. Instead of
building the acutumine core from a six-membered ring com-
pound, we decided to make the pyrrolidine ring (highlighted
in 1) the foundational element of our synthesis. This new
plan had the advantage that keto proline esters of type 4
are readily available by the intramolecular carbenoid inser-
tion chemistry of Rapoport and co-workers10 and amenable
to a range of chemical reactions. From this vantage point, we
would attempt a synthesis of tricyclic b-keto ester 5, which
emerged as an attractive subgoal for our effort. The manner
in which a compound embodying the architecture of 5 was
fashioned from pyrrolidine derivative 4 is described below.
While such a transformation could well be mechanistically
degenerate,11 we envisioned that it might occur by a se-
quence of carbonyl-dependent reactions starting with a base-
mediated b-elimination (6/7). This event would leave in
its wake a delocalized enolate ion and an electrophilic
enone, thus enabling an intramolecular Michael reaction to
give compound 8.12 An analogous rearrangement carried
13
€
out by Vorbruggen and co-workers bolstered our confi-
dence in this type of crowded C–O/C–C bond isomeriza-
tion. In base, enolization of the newly formed b-keto ester
would render this grouping unreactive toward nucleophiles.
A subsequent kinetic enolization of the methyl ketone in 8
would then trigger a final Dieckmann-like cyclization14
to tricyclo[4.3.3.0]dodecane 9, a compound having key ele-
ments of acutumine. The synthesis and reactivity of bicyclic
vinylogous carbonate 6 are described below.
O
3. Results and discussion
N
Boc
Steps
OH
O
O
As noted earlier, the functionalized pyrrolidine derivative 4
(wherein P¼Boc) is readily available by the method of
Rapoport and co-workers;10 this method was modified
slightly15 and applied to a synthesis of multi-gram quantities
of this useful building block. Alkylation of the stabilized
sodium enolate derived from 4 with propargyl bromide in
a mixed solvent system of THF and DMF afforded com-
pound 10 in racemic form in 71% yield (only one enantiomer
is shown for clarity) (Scheme 3). Compound 11 was subse-
quently produced by a chemoselective addition of 2-methyl-
allylmagnesium chloride to the ketone carbonyl of 10. This
reaction was efficient and highly diastereoselective; nucleo-
philic addition occurred to the face away from the propargyl
chain.16 A Sharpless, vanadium-based epoxidation17 of
the disubstituted alkene function in 11 was also highly dia-
stereoselective. This oxidation, which was presumably
hydroxyl-directed,18 gave rise to essentially only one diaste-
reoisomer, although we did not establish the configuration of
the newly formed stereocenter. This stereochemical matter
was ultimately inconsequential because we elected to carry
the epoxide function as a direct precursor to the required
trigonal keto group after several unsuccessful attempts at
achieving a direct oxidative cleavage of the alkene in
compound 11.
O
O
2: p-Benzoquinone
3
?
H3CO
OH
Cl
O
N
CH3
OCH3
OCH3
O
?
O
O
1: Acutumine
H3CO
O
Steps
N
P
N
P
MeO2C
O
OCH3
OCH3
5
4
Scheme 1. The issue of starting material for a synthesis of acutumine (1).
Our intention was to achieve a spontaneous conversion of
bicyclic vinylogous carbonate 6 to the tricyclic core of
acutumine in the presence of a suitable base (Scheme 2).
CO2Me
CO2Me
O
O
O
O
With hydroxyl and alkyne groupings in neighboring regions
of space, compound 12 was an excellent substrate for a pal-
ladium(II)-catalyzed, carbonylative cyclization19 to mixed,
vinylogous carbonate 13. This transformation was both rapid
and highly efficient (consistently >90% crude yield).
Finally, on exposure to periodic acid buffered with sodium
periodate,20 the epoxide ring in 13 was cleanly converted
to methyl ketone 6, a compound that was designed to be a
direct precursor to the core structure of acutumine (Scheme
2). In fact, it was possible to directly convert bicyclic vinyl-
ogous carbonate 6 to tricycle 9 on treatment of 6 with strong,
non-nucleophilic bases, although these reactions were in-
efficient and produced several byproducts. The preferred
course of action is shown below in Scheme 4. Under rather
gentle reaction conditions,21 compound 6 was smoothly
transformed to bicyclic polycarbonyl 8, almost certainly
via a b-elimination/Michael cyclization mechanism. In rela-
tion to compound 6, which proved to be somewhat sensitive,
Base
Elimination
H3C
N
N
H3C
Boc
Boc
H3CO
O
H3CO
O
6
7
Intramolecular
Michael reaction
O
O
O
O
Dieckmann-like
cyclization
H3CO
H3CO
N
N
Boc
Boc
CO2Me
CH3
O
OH
O
9
8
Scheme 2. A concept for generating the acutumine tricyclic core via a
cascade of carbonyl-dependent reactions. Boc¼t-BuOCO.

6448
R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
OH
O
O
O
b
a
OCH3
Boc
N
Boc
N
Boc
N
MeO2C
MeO2C
4
10
11
c
CO2Me
CO2Me
O
O
O
O
O HO
d
e
N
N
N
H3C
H3C
H3C
Boc
Boc
Boc
H3CO
O
H3CO
O
H3CO
O
6
13
12
Scheme 3. Synthesis of bicyclic vinylogous carbonate 6: (a) sodium hydride, THF/DMF (9:1), 0 ꢀC/rt, then propargyl bromide, 71% yield; (b) 2-methyl-
allylmagnesium chloride, THF, ꢁ78/0 ꢀC, 91% yield; (c) VO(acac)2, t-BuOOH, CH2Cl2, 0 ꢀC/rt, 80% yield; (d) (CH3CN)2PdCl2 (5 mol %), benzoquinone
(1.1 equiv), CO (balloon), MeOH, 0 ꢀC, 90% crude yield; (e) NaIO4, H5IO6, THF/H2O, 0 ꢀC/rt. THF¼tetrahydrofuran; DMF¼N,N-dimethylformamide;
acac¼acetyl acetonate.
CO2Me
O
O
O
O
O
O
H3CO
H3CO
a
b
N
N
N
H3C
Boc
Boc
CO2Me
CH3
Boc
H3CO
O
O
O
OH
6
8
9
Scheme 4. Construction of the core structure of acutumine: (a) n-Bu4NOAc, THF, rt, 64% yield, two steps; (b) NaN(SiMe3)2 (3.0 equiv), THF, ꢁ78 ꢀC/rt,
84% yield.
compound 8 is a stable substance that could be purified
by silica gel chromatography.22 When a solution of 8 in THF
was treated with 3.0 equiv of sodium bis(trimethylsilyl)-
amide under the conditions shown, the desired Die-
ckmann-like cyclization occurred and generated compound
5. Experimental
5.1. General
Unless otherwise noted, all reactions were carried out under
an atmosphere of argon or nitrogen. Tetrahydrofuran (THF),
toluene, diethyl ether, methylene chloride (CH2Cl2), and
N,N-dimethylformamide (DMF) were dried by passing
through activated alumina columns. Commercial reagents
of high purity were purchased and used without further pu-
rification unless otherwise noted. Reactions were monitored
by thin-layer chromatography (TLC) carried out on 0.25 mm
Merck silica gel plates (60 F254) using UV light as a visual-
izing agent and aqueous ceric sulfate/phosphomolybdic acid
or aqueous potassium permanganate solution and heat as
developing agents. E. Merck silica gel 60 (230–400 mesh)
was used for flash column chromatography. Instrumentation:
FTIR spectra were obtained on a Perkin–Elmer Paragon 500
FTIR spectrometer. NMR spectra were obtained on a Varian
Inova-500 instrument and calibrated to the residual solvent
peak. The multiplicities are abbreviated as follows:
s¼singlet, d¼doublet, t¼triplet, q¼quartet, m¼multiplet,
br¼broad signal. Mass spectra were obtained on an Agilent
ESI-TOF mass spectrometer.
1
9. The H NMR spectrum of 9 at room temperature in
CDCl3 is consistent with the structure and enolic form as
drawn. Sharp singlets for the signals corresponding to both
the 1,3-dicarbonyl enol hydrogen and the Boc tert-butyl hy-
drogens suggest an intramolecular hydrogen bond between
this enol hydrogen and the carbonyl of the Boc group.
1
This results in an overall simplification of the H and 13C
NMR spectra consistent with a single N-Boc rotamer.
4. Conclusion
It has been our goal to evolve an effective synthesis of
the complex topology of the well-known alkaloid acutu-
mine. To this end, we have produced the propellane-
like [4.3.3.0] fused tricyclic core of acutumine in only
seven transformations from a known and readily available
pyrrolidine derivative. Four of these transformations were
enabled by the properties and reactivity of the carbonyl
group. The concise construction of this rigid, acutumine-
like substructural element provides a basis for addressing
the larger goal of achieving a stereocontrolled synthesis
of the highly challenging full structure of acutumine, as
well as many other acutumine-like compounds. These ef-
forts are currently underway and will be described in due
course.

R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
6449
5.1.1. Masamune–Claisen condensation on N-Boc-b-ala-
nine. Under a blanket of inert gas, solid N-Boc-b-alanine
(24.31 g, 128.5 mmol, 1.0 equiv) was added in one portion
suspension of 1,10-carbonyldiimidazole (25 g,
to
a
154.2 mmol, 1.2 equiv) in 350 mL THF in a 1 L round bot-
tom flask via a powder funnel, which was rinsed with
25 mLTHF. The resulting suspension was allowed to stir un-
der a stream of inert gas for 1 h, during which time carbon
dioxide evolved and the reaction mixture became a homoge-
neous, yellow solution, then a septum and argon balloon
were placed on the flask. The yellow reaction solution was
allowed to stir 3 h further before a mixture of MgCl2 (325
mesh, 12.23 g, 128.5 mmol, 1.0 equiv) and methyl potas-
sium malonate (40.13 g, 257.0 mmol, 2.0 equiv) solids
were added under a blanket of inert gas in one portion via
a powder funnel, which was rinsed with 25 mL THF. A
septum and argon balloon were placed on the flask and the
suspension was allowed to stir vigorously overnight (slow
gas evolution) before 250 mL water and 150 mL 1 M HCl
were added. The resulting yellow, biphasic solution (aque-
ous layer pH approx 7) was poured into a 2 L separatory
funnel, followed by 1 L ethyl acetate, using some to rinse
the reaction flask. The aqueous layer was removed and the
organic layer was washed with 150 mL 1 M HCl, 150 mL
brine, dried with MgSO4, filtered, and concentrated under
reduced pressure to give, after azeotropic removal of the re-
maining ethyl acetate with three 200 mL portions of hexane,
33 g of the desired b-keto ester as a faint yellow liquid,
which was taken on without further purification. TLC:
Rf¼0.24 (SiO2, 1:2 ethyl acetate/hexane); IR (film) 3397,
2978, 1748, 1714, 1520, 1439, 1367, 1324, 1252,
5.1.3. Diazo transfer reaction on Masamune–Claisen b-
keto ester. Under a blanket of inert gas, the above sulfonyl
azide (32.11 g, 141.3 mmol, 1.1 equiv) was added in one por-
tion to a solution of the above b-keto ester (33 g, 128.5 mmol,
1.0 equiv) in 450 mL acetonitrile in a 1 L round bottom flask
via a powder funnel, which was rinsed with 50 mL aceto-
nitrile. Triethylamine (54 mL, 385.4 mmol, 3.0 equiv) was
added dropwise via a pressure equalizing addition funnel
over 30 min, and the resulting yellow-orange solution was
allowed to stir for 1 h before the solvent was removed under
reduced pressure. To the foamy, yellow-orange residue was
added a stirbar and 800 mL 1:1 diethyl ether/water. After
allowing the mixture to stir well for several minutes, the
resulting biphasic solution was poured into a 2 L separatory
funnel and the flask was rinsed with 500 mL 4:1 diethyl
ether/water and 400 mL diethyl ether. The aqueous layer
was removed and the organic layer was washed with
300 mL 2:1 saturated NaHCO3 solution/water and 300 mL
2:1 saturated NH4Cl solution/water. The organic layer was
dried with MgSO4, filtered, and concentrated under reduced
pressure to give 35 g of the desired diazo b-keto ester as a
viscous yellow liquid that was taken on without further puri-
fication. TLC: Rf¼0.31 (SiO2, 1:2 ethyl acetate/hexane); IR
1
(film) 3390, 2978, 2139, 1715, 1652, 1514, 1438 cmꢁ1; H
1
1170 cmꢁ1; H (500 MHz, CDCl3) d 5.05 (br s, 1H), 3.62
(500 MHz, CDCl3) d 5.03 (br s, 1H), 3.77 (s, 3H), 3.39 (s,
2H), 3.37 (q, J¼5.9 Hz, 2H), 2.97 (t, J¼5.9 Hz, 2H), 1.35
(s, 9H); 13C (125 MHz, CDCl3) d 191.5, 161.4, 155.6, 78.9,
76.0, 52.1, 40.6, 35.3, 28.2; HRMS (ESI-TOF) C11H17N3O5
m/z calcd for [M+Na]+ 294.1060; found 294.1056.
(s, 3H), 3.38 (s, 2H), 3.26 (q, J¼5.9 Hz, 2H), 2.68 (t, J¼
5.9 Hz, 2H), 1.31 (s, 9H); 13C (125 MHz, CDCl3) d 202.0,
167.1, 155.6, 79.0, 52.1, 48.8, 42.8, 34.8, 28.1; HRMS
(ESI-TOF) C11H19NO5 m/z calcd for [M+Na]+ 268.1155;
found 268.1151.
1
The H and 13C spectra of all compounds below, with the
exception of the final compound (9), are complicated by
N-Boc rotamers.
5.1.2. 3-Carboxybenzenesulfonyl azide. 3-(Chlorosulfo-
nyl)benzoic acid (50 g, 226 mmol, 1.0 equiv) was added in
one portion to 660 mL acetone in a 2 L round bottom flask
via a powder funnel, which was rinsed with 50 mL acetone.
To the resulting light brown solution was added a solution of
sodium azide (18.4 g, 284 mmol, 1.25 equiv) in 120 mL
water, followed by approximately 20–30 mL water to
make the reaction mixture homogeneous. The light brown
solution was allowed to stir for 2 h before most of the ace-
tone was removed under reduced pressure leaving a brown
residue. Approximately 1500 mL of water was added to the
flask, followed by a few milliliters of concentrated HCl
(solution pH <2). The precipitate was collected in a
150 mL sintered glass funnel, rinsed well with water and
three portions of hexanes in the funnel, then dried by pulling
air through the funnel for 1 h followed by drying under high
vacuum overnight to give 41 g of the desired sulfonyl azide
as an off-white, powdery solid (80%). IR (film) 3100–2200,
O
O
O
O
Rh2(OAc)4
Boc
OCH3
Boc
N
H
OCH3
toluene, 85-90 °C
N
N2
4
5.1.4. N-Boc-3-keto proline methyl ester 4. Rh2(OAc)4
(284 mg, 0.642 mmol, 0.005 equiv) was added to a pale
yellow solution of the above diazo b-keto ester (34.9 g,
128.5 mmol, 1.0 equiv) in 1300 mL of toluene in a 2 L round
bottom flask, which was then fitted with a reflux condenser
and placed into a preheated 100 ꢀC oil bath under a stream
of inert gas (no septum). The green mixture was allowed
to heat at 85–90 ꢀC until nitrogen gas evolution ceased (ap-
proximately 30 min) and then was allowed to cool to room
temperature. The solvent was removed under reduced pres-
sure, then under high vacuum, and the green residue was
then treated with 500 mL hexane and the solvent was again
removed under reduced pressure and high vacuum. To
remove the Rh2(OAc)4, the green residue was diluted with
diethyl ether and filtered through CeliteÒ, rinsing well
1
2135, 1913, 1712, 1374, 1265, 1177 cmꢁ1; H (500 MHz,
CDCl3/DMSO-d6) d 11.41 (br s, 1H), 8.58 (t, J¼1.6 Hz,
1H), 8.35 (ddd, J¼7.8, 1.5, 1.3 Hz, 1H), 8.07 (ddd, J¼7.9,
2.0, 1.2 Hz, 1H), 7.67 (t, J¼7.9 Hz, 1H); 13C (125 MHz,
CDCl3/DMSO-d6) d 166.1, 138.7, 135.6, 132.9, 130.8,
129.8, 128.6; HRMS (ESI-TOF) C7H5N3O4S m/z calcd for
[MꢁH]ꢁ 225.9928; found 225.9926.

6450
R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
with diethyl ether, and the solvent was evaporated under
reduced pressure to give 29 g of a yellow oil (93%), which
solidified when azeotroped with hexanes or placed in the
freezer. This material is sufficiently pure to be taken on with-
out any further purification, but can be purified by stirring
the crude product with 1:1 diethyl ether/hexane, collecting
the solid by filtration, evaporation of the filtrate, and repeat-
ing the sequence. An analytical sample was obtained by flash
column chromatographic purification (SiO2, 1:3 ethyl ace-
tate/hexane) to give a colorless powder, mp 55–56 ꢀC.
TLC: Rf¼0.32 (SiO2, 1:2 ethyl acetate/hexane); IR (film)
extracts dried with MgSO4, and the solvent evaporated under
reduced pressure. The residue was purified by flash chroma-
tography (SiO2, 1:4:1 diethyl ether/hexanes/dichloro-
methane/1:3:1/1:2:1/1:1:1) to give 1.09 g of 11 as a
colorless solid (91%), mp 77–79 ꢀC. TLC: Rf¼0.45 (SiO2,
1:2 ethyl acetate/hexane); IR (film) 3462, 3308, 2977,
2250, 2122, 1739, 1694, 1394, 1258, 1172 cmꢁ1 1H
;
(500 MHz, CDCl3) d 4.90 (s, 1H), 4.74 (s, 1H), 3.82–3.72
and 3.72–3.58 (2 m, 4H total), 3.67 (s, 3H), 3.56–3.39 (m,
1H), [3.16 and 3.04 (2 dd, J¼17.0, 2.3 Hz), 3.02 and 2.95
(2 dd, J¼17.0, 2.6 Hz) 2H total], 2.72 (d, J¼15.5 Hz, 1H),
2.56 and 2.46 (2 d, J¼13.4 Hz, 1H total), 2.14–2.02 (m,
2H), 1.99–1.91 (m, 1H), 1.88 (d, J¼13.7 Hz, 1H), 1.79 (s,
3H), 1.40 and 1.37 (2 s, 9H total); 13C (125 MHz, CDCl3)
d 172.5, 172.3, 154.2, 153.4, 141.6, 141.4, 116.3, 116.1,
84.9, 83.9, 82.1, 81.6, 80.8, 80.2, 73.7, 73.5, 71.2, 71.0,
52.5, 45.9, 45.6, 44.2, 36.1, 35.5, 28.6, 28.4; HRMS (ESI-
TOF) C18H27NO5 m/z calcd for [M+Na]+ 360.1781; found
360.1778.
2977, 1775, 1746, 1707, 1395, 1239, 1163 cmꢁ1 1H
;
(500 MHz, CDCl3) d 4.55 and 4.47 (2 s, 1H total), 3.93–
3.75 (m, 5H), 2.67 (t, J¼7.6 Hz, 2H), 1.47 and 1.40 (2 s,
9H total); 13C (125 MHz, CDCl3) d 204.6, 166.7, 153.7,
81.1, 65.6, 65.2, 52.9, 42.1, 41.5, 37.1, 36.4, 28.1; HRMS
(ESI-TOF) C11H17NO5 m/z calcd for [M+Na]+ 266.0999;
found 266.0990.
5.1.5. Propargyl keto proline 10. A suspension of sodium
hydride (60% dispersion in mineral oil, 181 mg,
4.52 mmol, 1.1 equiv) in 16 mL THF and 2 mL DMF was
cooled to 0 ꢀC, then a solution of keto proline 4 (1 g,
4.11 mmol, 1.0 equiv) in 2 mL THF was added over 5–
10 min, followed by a rinse with 1 mL THF. The resulting
yellow suspension was allowed to stir at 0 ꢀC for 30 min be-
fore removing the flask from the ice bath. Propargyl bromide
(80 wt % in toluene, 440 mL, 4.93 mmol, 1.2 equiv) was
then added dropwise over 5 min and the resulting orange-
red suspension was allowed to stir at room temperature for
30 min before adding 5 mL saturated NH4Cl solution and
10 mL water. The biphasic solution was poured into a
250 mL separatory funnel containing 100 mL water, the
organic layer was removed, and the aqueous phase was
extracted 4ꢂ25 mL diethyl ether. The combined organic
extracts were dried with MgSO4, filtered, and the solvent
removed under reduced pressure. The residue was purified
by flash chromatography (SiO2, 1:6:1 diethyl ether/hexanes/
dichloromethane/1:4:1/1:2:1) to give 820 mg of 10 as
small, colorless beads (71%), mp 84–86 ꢀC. TLC: Rf¼0.40
(SiO2, 1:2 ethyl acetate/hexane); IR 3289, 2978, 2360,
5.1.7. Epoxide 12. To a solution of homoallylic alcohol 11
(1 g, 2.96 mmol, 1.0 equiv) in 30 mL CH2Cl2 was added
VO(acac)2 and the resulting green solution was cooled to
0 ꢀC before tBuOOH (5.5 M in decane, 1.62 mL,
8.89 mmol, 3.0 equiv) was added dropwise. After allowing
the reddish purple solution to stir at 0 ꢀC for 15 min, the
solution was allowed to warm to room temperature and stir
there overnight. The resulting yellow-orange solution was
poured into a 125 mL separatory funnel, the flask was rinsed
with 30 mL of CH2Cl2, 20 mL brine was added, the pale
yellow organic layer was removed, and the aqueous layer
was extracted 2ꢂ10 mL CH2Cl2. The combined organic
layers were dried over MgSO4, filtered, and the solvent re-
moved under reduced pressure. The residue was purified
by flash chromatography (SiO2, 1:4:1 diethyl ether/hexanes/
dichloromethane/1:3:1/1:2:1/1:1:1) to give 832 mg of
12 as a colorless oil (80%). When this oil was treated with
a small amount of diethyl ether a colorless solid formed,
which gave way to a colorless foam upon evaporation of
the solvent. By scratching the sides of the flask, 12 could
be isolated as a colorless powder, mp 83–88 ꢀC. TLC:
Rf¼0.38 (SiO2, 1:1 ethyl acetate/hexane); IR (film) 3461,
1
1774, 1743, 1706, 1386, 1255, 1147 cmꢁ1; H (500 MHz,
CDCl3) d 3.97–3.89 and 3.89–3.77 (2 m, 2H total), 3.71 (s,
3H), 3.35 and 3.11 (dd and qd, J¼17.0, 2.6 Hz, 2H total),
2.86–2.64 (m, 2H), 1.97 and 1.94 (2 t, J¼2.6 Hz, 1H total),
1.49 and 1.43 (2 s, 9H total); 13C (125 MHz, CDCl3) d 206.9,
206.5, 167.4, 154.0, 153.4, 81.6, 81.1, 78.9, 78.2, 71.9, 71.6,
71.5, 70.9, 53.1, 42.8, 42.2, 36.5, 35.7, 28.3, 28.2; HRMS
(ESI-TOF) C14H19NO5 m/z calcd for [M+Na]+ 304.1155;
found 304.1155.
3307, 2977, 1737, 1696, 1394, 1256, 1172 cmꢁ1 1H
;
(500 MHz, CDCl3) d 3.89–3.82, 3.76–3.64 and 3.62–3.42
(3 m, 6H total), 3.25–2.91 (m, 2H), 2.76–2.54 (m, 2H),
2.34–1.95 (m, 4H), 1.48–1.28 (m, 12H); 13C (125 MHz,
CDCl3) d 172.4, 154.2, 153.4, 85.4, 84.3, 81.9, 81.4, 80.9,
80.3, 73.9, 73.7, 71.1, 70.8, 56.0, 55.6, 52.6, 46.1, 45.8,
41.7, 36.5, 35.9, 28.6, 28.4; HRMS (ESI-TOF) C18H27NO6
m/z calcd for [M+Na]+ 376.1730; found 376.1716.
5.1.6. Homoallylic alcohol 11. To a solution of propargyl
keto proline 10 (1 g, 3.55 mmol, 1.0 equiv) in 10.5 mL
THF cooled to ꢁ78 ꢀC was added 2-methylallylmagnesium
chloride (0.5 M in THF, 7.11 mL, 1.0 equiv) dropwise over
15–20 min via syringe pump. The resulting colorless solu-
tion was allowed to slowly warm to between ꢁ10 ꢀC and
0 ꢀC over 90 min, after which 30 mL saturated NH4Cl solu-
tion was added and the mixture was allowed to warm to room
temperature. The biphasic solution was poured into a
125 mL separatory funnel, 10 mL diethyl ether was added,
and the organic layer was removed. The aqueous phase
was extracted 4ꢂ15 mL diethyl ether, the combined organic
5.1.8. Vinylogous carbonate 13. To a three-neck 250 mL
round bottom flask containing 80 mL of anhydrous methanol
was added freshly sublimed yellow needles of 1,4-benzo-
quinone (475 mg, 4.39 mmol, 1.1 equiv), followed by
(MeCN)2PdCl2 (52 mg, 0.199 mmol, 0.05 equiv) under a
stream of inert gas delivered via a three-way adapter attached
to the central neck of the flask. A balloon of carbon monoxide
was attached to the three-way adapter and the flask was evac-
uated and refilled with carbon monoxide three times. While
maintaining the reaction under 1 atm of carbon monoxide,
the orange solution was allowed to cool to 0 ꢀC before a solu-
tion of epoxide 12 (1.41 g, 3.99 mmol, 1.0 equiv) in 36 mL

R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
6451
methanol was added dropwise over several minutes, followed
by a rinse with 4 mL methanol. The resulting solution was
allowed to stir at 0 ꢀC for 30 min before pouring the orange
solution into a 2 L separatory funnel. The flask was rinsed
3ꢂ200 mL CH2Cl2, and then 800 mL 5% NaOH was added.
The organic layer was removed, and the brown/black aqueous
layer was extracted 3ꢂ100 mL CH2Cl2. The combined or-
ganic layers were dried over MgSO4, filtered, and the solvent
removed under reduced pressure to give 1.48 g of the desired
product as an off-white foam (90%). This material was taken
on without further purification as it partially decomposed
when subjected to purification by flash chromatography on
silica gel. However, an analytical sample was obtained by
flash column chromatographic purification (SiO2, 1:1 diethyl
ether/hexane/1:1:1 diethyl ether/hexane/dichloromethane)
to give a colorless foam. TLC: Rf¼0.42 (SiO2, 1:1 ethyl ace-
tate/hexane); IR (film) 3482, 2980, 2953, 1747, 1699, 1652,
resulting yellow suspension was allowed to stir at room tem-
perature for 30 min before removing most of the solvent under
reduced pressure. To the residue was added 13 mL water,
2.6 mL 1 M NaOH, and 3 mL hexane. The resulting
biphasic solution was poured into a 30 mL separatory funnel,
the organic layer was removed, and the aqueous layer was
washed 3ꢂ3 mL hexane. To the aqueous layer was then added
10% HCl until the solution was acidic and the aqueous layer
was then extracted 5ꢂ15 mL diethyl ether. The combined
diethyl ether extracts were dried over MgSO4, filtered, and
the solvent removed under reduced pressure. The residue
was purified by flash chromatography (SiO2, 30% diethyl
ether/hexane/40%/50%/60%/70%, all eluant con-
taining 1% AcOH) to give 171 mg of 8 as a colorless oil
(64%, two steps). TLC: Rf¼0.34 (SiO2, 1:1 ethyl acetate/hex-
ane+1% AcOH); IR (film) 2978, 2954, 1749, 1701, 1663,
1
1618, 1448, 1393 cmꢁ1; H (500 MHz, CDCl3) d 10.76 (br
1
1437, 1394 cmꢁ1; H (500 MHz, CDCl3) d 5.32 (m, 1H),
s, 1H), 3.83–3.63 (m, 9H), 3.22–3.13 (m, 2H), 2.98 (d,
J¼18.8 Hz, 0.4H), 2.82 (d, J¼18.5 Hz, 0.6H), 2.78 and 2.77
(2 d, J¼15.9 and 15.8 Hz, 1.5H total), 2.53 and 2.49 (2 d,
J¼15.8 and 15.9 Hz, 1H total), 2.41–2.32 (m, 1H), 2.13–
1.96 (m, 5H), 1.41 and 1.38 (2 s, 9H total); 13C (125 MHz,
CDCl3) d 205.4, 205.2, 176.6, 175.8, 171.4, 168.9, 168.7,
153.4, 152.5, 99.3, 80.6, 80.2, 73.8, 73.6, 60.0, 58.7, 52.4,
52.3, 51.1, 51.0, 47.6, 47.5, 45.6, 45.4, 41.5, 40.4, 33.1,
32.6, 31.4, 28.2; HRMS (ESI-TOF) C19H27NO8 m/z calcd
for [M+H]+ 398.1809; found 398.1799.
3.96–3.69 (m, 3H), 3.71 (s, 3H), 3.64 (s, 3H), 3.43–3.22 (m,
2H), 2.71 (d, J¼4.6 Hz, 1H), 2.61 (d, J¼4.6 Hz, 1H), 2.41–
1.97 (m, 2H), 1.49–1.36 (m, 12H); 13C (125 MHz, CDCl3)
d 172.6, 172.4, 169.7, 168.4, 168.2, 153.5, 152.6, 98.1,
97.0, 92.3, 92.0, 81.5, 81.0, 78.7, 76.5, 54.8, 53.5, 53.2,
51.1, 45.6, 45.4, 41.1, 40.8, 39.4, 38.8, 33.5, 32.9, 28.5;
HRMS (ESI-TOF) C20H29NO8 m/z calcd for [M+H]+
412.1966; found 412.1951.
5.1.9. Ketone 6. A solution of crude vinylogous carbonate
13 (276 mg, 0.671 mmol, 1.0 equiv) in 6.7 mL THF and
670 mL water was cooled to 0 ꢀC, then NaIO4 (86 mg,
0.402 mmol, 0.6 equiv) and H5IO6 (184 mg, 0.805 mmol,
1.2 equiv) were added. The resulting solution was allowed
to stir at 0 ꢀC for 15 min, then at room temperature over-
night, during which time sodium iodate precipitated. The
resulting suspension was poured into a 60 mL separatory
funnel, followed by 3.5 mL saturated NaHCO3 solution,
7 mL water, and 10 mL diethyl ether. The organic layer
was removed and the aqueous layer was extracted
3ꢂ10 mL diethyl ether. The organic extracts were dried
over MgSO4, filtered, and the solvent removed under re-
duced pressure to give 270 mg of an off-white foam. This
material was taken on without further purification as it sig-
nificantly decomposed when subjected to purification by
flash chromatography on silica gel. However, an analytical
sample was obtained by flash column chromatographic
purification (SiO2, 1:1:1 diethyl ether/hexane/dichloro-
methane/2:1:1/diethyl ether) to give a colorless foam.
TLC: Rf¼0.39 (SiO2, 1:1 ethyl acetate/hexane); IR (film)
5.1.11. Tricycle 9. A solution of b-keto ester 8 (100 mg,
0.252 mmol, 1.0 equiv) in 5 mL THF was cooled to
ꢁ78 ꢀC, then NaHMDS (0.6 M in toluene, 1.26 mL,
0.755 mmol, 3.0 equiv) was added dropwise. The resulting
yellow solution was allowed to stir at ꢁ78 ꢀC for 15 min,
then at room temperature for 90 min. At room temperature,
3 mL 1 M NaHSO4 was slowly added to the yellow-orange
suspension, the organic layer was removed, and the aqueous
layer (pH <2) was extracted 4ꢂ1 mL ethyl acetate. The
combined organic extracts were washed 3ꢂ1 mL brine,
dried over Na2SO4, filtered, and the solvent removed under
reduced pressure. The residue was purified by flash chroma-
tography (SiO2, 30% diethyl ether/hexane/40%/50%/
60%/70%, all eluant containing 1% AcOH) to give,
after azeotropic removal of AcOH with three portions of
hexanes, 77 mg of 9 (84%) as a colorless powder, mp 135–
139 ꢀC. TLC: Rf¼0.34 (SiO2, 1:1 ethyl acetate/hexane+1%
AcOH); IR (film) 2980, 2716, 2360, 2341, 1772, 1734,
1717, 1661, 1447, 1406 cmꢁ1 1H (500 MHz, CDCl3)
;
d 11.41 (s, 1H), 10.65 (br s, 1H), 5.38 (s, 1H), 3.79 (s,
3H), 3.65 (d, J¼18.7 Hz, 1H), 3.46 (dd, J¼10.7, 8.3 Hz,
1H), 3.19 (ddd, J¼12.2, 10.7, 6.3 Hz, 1H), 2.96 and 2.93
(2 d, J¼18.7 and 16.5 Hz, 2H total), 2.41 (dd, J¼13.0,
6.2 Hz, 1H), 2.37 (d, J¼16.5 Hz, 1H), 1.82 (dt, J¼12.7,
12.6, 8.3 Hz, 1H), 1.47 (s, 9H); 13C (125 MHz, CDCl3)
d 196.1, 173.0, 172.7, 169.0, 158.5, 103.5, 103.3, 83.6,
67.7, 56.2, 51.6, 47.0, 44.6, 38.7, 31.0, 28.2; HRMS (ESI-
TOF) C18H23NO7 m/z calcd for [M+H]+ 366.1547; found
366.1533.
3472, 2979, 2954, 2901, 2254, 1780–1650 cmꢁ1
;
1H
(500 MHz, CDCl3) d 5.18 (m, 1H), 3.89–3.53 (m, 6H),
3.50 (s, 3H), 3.24–3.11 (m, 1H), 2.78 and 2.74 (2 d,
J¼14.2 and 14.9 Hz, 1H total), 2.36 (dd, J¼13.9, 5.9 Hz,
1H), 2.14 and 2.13 (2 d, J¼15.1 and 15.3 Hz, 1H total),
2.10–1.97 (m, 4H), 1.28 and 1.26 (2 s, 9H total); 13C
(125 MHz, CDCl3) d 203.3, 203.0, 171.9, 169.1, 167.7,
153.0, 152.0, 96.7, 95.6, 91.8, 91.5, 80.9, 80.4, 75.5, 75.3,
52.8, 50.5, 45.9, 45.8, 45.0, 44.7, 39.2, 38.5, 32.7, 32.1,
31.5, 27.9; HRMS (ESI-TOF) C19H27NO8 m/z calcd for
[M+Na]+ 420.1629; found 420.1622.
Acknowledgements
5.1.10. Bicyclic b-keto ester 8. Bu4NOAc (405 mg,
1.34 mmol, 2.0 equiv) was added to a solution of crude ketone
6 (270 mg, 0.671 mmol, 1.0 equiv) in 13 mL THF and the
This work was supported by the National Institute of
General Medical Sciences (GM074763), Merck Research

6452
R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
Winterfeldt, E. Chem.—Eur. J. 2003, 9, 3849–3858; For related
studies on oxygen-to-carbon [1,3]-shifts, see: (d) Trost, B. M.;
Runge, T. A. J. Am. Chem. Soc. 1981, 103, 7550–7559; (e)
Swenton, J. S.; Bradin, D.; Gates, B. D. J. Org. Chem. 1991,
56, 6156–6163; (f) Swenton, J. S.; Callinan, A.; Wang, S.
J. Org. Chem. 1992, 57, 78–85.
Laboratories, Boehringer Ingelheim, a Johnson & Johnson
Focused Giving Award (E.J.S.), Princeton University, and
a predoctoral fellowship from Bristol-Myers Squibb
€
(R.J.M.). We thank Dr. Gunther Scheid for his early efforts
on this project, particularly on the quinone-based route.
ꢀ
We also thank Dr. Istvan Pelczer and Dr. Carlos Pacheco
14. For a creative application of tandem Dieckmann cyclizations
in complex chemical synthesis, see: (a) Stork, G.; LaClair,
J. J.; Spargo, P.; Nargund, R. P.; Totah, N. J. Am. Chem.
Soc. 1996, 118, 5304–5305; For a review of the Dieckmann
condensation and related processes, see: (b) Davis, B. R.;
Garratt, P. J. Comprehensive Organic Synthesis: Additions
to C–X p-Bonds, Part 2; Trost, B. M., Fleming, I., Eds.;
Pergamon: New York, NY, 1991; Vol. 2, Chapter 3.6,
pp 795–863; (c) De Risi, C.; Spalluto, G.; Zanirato, V.
Chem. Rev. 1995, 95, 1065–1114.
15. A one-pot Masamune–Claisen condensation was used. (a)
Brooks, D. W.; Lu, L. D.-L.; Masamune, S. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 72–74; (b) Smith, T. E.; Djang, M.;
Velander, A. J.; Downey, C. W.; Carroll, K. A.; van Alphen,
S. Org. Lett. 2004, 6, 2317–2320. In place of 4-carboxybenz-
enesulfonyl azide for the diazo transfer reaction, the less ex-
pensive 3-carboxybenzenesulfonyl azide was used. For the
insertion reaction, toluene was used rather than benzene. See
Section 5 for details.
(Princeton) for NMR assistance, and Dr. Gary Siuzdak
(The Scripps Research Institute) for mass spectral analysis.
References and notes
1. (a) Robinson, R. J. Chem. Soc., Trans. 1917, 111, 762–768; (b)
Gupta, A. K.; Fu, X.; Snyder, J. P.; Cook, J. M. Tetrahedron
1991, 47, 3665–3710; (c) Ho, T. L. Tandem Organic
Reactions; John Wiley & Sons: New York, NY, 1992; (d)
Wasserman, H. H.; Parr, J. Acc. Chem. Res. 2004, 37,
687–701.
2. Yu, B.-W.; Chen, J.-Y.; Wang, Y.-P.; Cheng, K.-F.; Li, X.-Y.;
Qin, G.-W. Phytochemistry 2002, 61, 439–442.
3. Qin, G.-W.; Tang, X.-C.; Caignard, D.-H.; Renard, P.; Lestage,
P. Acutumine and Acutumine Compounds, Synthesis and Use.
Patent number WO2004000815, International publication date
Dec 31, 2003.
16. (a) Molander, G. A.; Harris, C. R. J. Am. Chem. Soc. 1995, 117,
3705–3716; (b) Eicher, T.; Servet, F.; Speicher, A. Synthesis
1996, 863–870; (c) Molander, G. A.; Harris, C. R. J. Org.
Chem. 1997, 62, 2944–2956; (d) Chen, C.; Layton, M. E.;
Shair, M. D. J. Am. Chem. Soc. 1998, 120, 10784–10785 and
references cited therein.
17. (a) Sharpless, K. B.; Michaelson, R. C. J. Am. Chem. Soc. 1973,
95, 6136–6137; (b) Sharpless, K. B.; Verhoeven, T. R.
Aldrichimica Acta 1979, 12, 63–74.
4. Goto, K.; Sudzuki, H. Bull. Chem. Soc. Jpn. 1929, 4, 220–
224.
5. (a) Goto, K.; Tomita, M.; Okamoto, Y.; Sasaki, Y.; Matoba, K.
Proc. Jpn. Acad. 1966, 42, 1181–1184; (b) Tomita, M.;
Okamoto, Y.; Kikuchi, T.; Osaki, K.; Nishikawa, M.;
Kamiya, K.; Sasaki, Y.; Matoba, K.; Goto, K. Tetrahedron
Lett. 1967, 25, 2421–2424; (c) Tomita, M.; Okamoto, Y.;
Kikuchi, T.; Osaki, K.; Nishikawa, M.; Kamiya, K.; Sasaki,
Y.; Matoba, K.; Goto, K. Tetrahedron Lett. 1967, 25, 2425–
2430; (d) Goto, K.; Tomita, M.; Okamoto, Y.; Kikuchi, T.;
Osaki, K.; Nishikawa, M.; Kamiya, K.; Sasaki, Y.; Matoba,
K. Proc. Jpn. Acad. 1967, 43, 499–504.
18. For an excellent review on ‘substrate-directable chemical reac-
tions’, see: Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev.
1993, 93, 1307–1370.
19. Intramolecular alkoxycarbonylation of alkynes: (a) Gabriele,
B.; Salerno, G.; De Pascali, F.; Costa, M.; Chiusoli, G. P.
J. Organomet. Chem. 2000, 593–594, 409–415; (b) Marshall,
J. A.; Yanik, M. M. Tetrahedron Lett. 2000, 41, 4717–4721;
(c) Kato, K.; Nishimura, A.; Yamamoto, Y.; Akita, H.
Tetrahedron Lett. 2001, 42, 4203–4205; Intramolecular
alkoxycarbonylation of alkenes: (d) Semmelhack, M. F.;
Bodurow, C. J. Am. Chem. Soc. 1984, 106, 1496–1498; (e)
Semmelhack, M. F.; Kim, C.; Zhang, N.; Bodurow, C.;
Sanner, M.; Dobler, W.; Meir, M. Pure Appl. Chem. 1990,
62, 2035–2040; Intermolecular alkoxycarbonylation of al-
kenes: (f) James, D. E.; Stille, K. J. Am. Chem. Soc. 1976,
98, 1810–1823.
20. Presumably due to the high acidity of H5IO6 (pKa’s 1.6, 5.4,
Fieser & Fieser), we and others have found it useful to carry
out reactions of this type under buffered conditions such as
NaIO4/AcOH or NaIO4/H5IO6. (a) Nagarkatti, J. P.; Ashley,
K. R. Tetrahedron Lett. 1973, 46, 4599–4600; (b) Marko,
I. E.; Plancher, J.-M. Tetrahedron Lett. 1999, 40, 5259–
5262; (c) Evarts, J. B., Jr.; Fuchs, P. L. Tetrahedron Lett.
2001, 42, 3673–3675; (d) Snyder, S. A.; Corey, E. J. J. Am.
Chem. Soc. 2006, 128, 740–742; Others have exploited this
acidity: (e) Okamoto, S.; Shimazaki, T.; Kitano, Y.;
Kobayashi, Y.; Sato, F. J. Chem. Soc., Chem. Commun.
1986, 1352–1355; (f) Wu, W.-L.; Wu, Y.-L. J. Org. Chem.
1993, 58, 3586–3588.
6. Barton, D. H. R.; Kirby, A. J.; Kirby, G. W. J. Chem. Soc. C
1968, 929–936.
7. In nature, acutumine originates from two molecules of tyrosine.
For biosynthetic studies, see: (a) Sugimoto, Y.; Uchida, S.;
Inanaga, S.; Kimura, Y.; Hashimoto, M.; Isogai, A. Biosci.
Biotechnol. Biochem. 1996, 60, 503–505; (b) Babiker,
H. A. A.; Sugimoto, Y.; Saisho, S.; Inanaga, S.; Hashimoto,
M.; Isogai, A. Biosci. Biotechnol. Biochem. 1999, 63,
515–518.
8. (a) Matoba, K.; Karibe, N.; Yamazaki, T. Chem. Pharm. Bull.
1984, 32, 2639–2645; (b) Waller, D. L.; Stephenson, C. R. J.;
Wipf, P. Org. Biomol. Chem. 2007, 5, 58–60.
9. Reeder, M. D.; Srikanth, G. S. C.; Jones, S. B.; Castle, S. L.
Org. Lett. 2005, 7, 1089–1092.
10. Moyer, M. P.; Feldman, P. L.; Rapoport, H. J. Org. Chem. 1985,
50, 5223–5230.
11. Hudlicky, T. Chem. Rev. 1996, 96, 3–30.
12. (a) Bergmann, E. D.; Ginsburg, D.; Pappo, R. Org. React. 1959,
10, 179–555; (b) Little, R. D.; Masjedizadeh, M. R.; Wallquist,
O.; McLoughlin, J. I. Org. React. 1995, 47, 315–552; (c) Ihara,
M.; Fukumoto, K. Angew. Chem., Int. Ed. Engl. 1993, 32,
1010–1022.
13. (a) Skuballa, W.; Vorbr€uggen, H. Angew. Chem. 1981, 93,
1080–1081; (b) Weyerstahl, P.; Meisel, T.; Mewes, K.;
Negahdari, S. Liebigs Ann. Chem. 1991, 1, 19–25; (c)
Knappwost-Gieseke, C.; Nerenz, F.; Wartchow, R.;

R. J. Moreau, E. J. Sorensen / Tetrahedron 63 (2007) 6446–6453
6453
21. We also found sodium hydroxide, tetrabutylammonium fluo-
ride, and cesium carbonate suitable to effect this and related
transformations. Surprisingly, amine bases (e.g., triethylamine,
DBU, etc.) were ineffective under identical conditions without
the addition of a mild Lewis acid such as lithium chloride.
22. Compounds containing the vinylogous carbonate moiety (13
and 6) proved sensitive toward purification on silica gel, though
they were of sufficient purity after aqueous workup to be taken
on into the next reaction. Purification via silica gel chromato-
graphy was carried out on compound 8.