Synthesis of polyprenylated acylphloroglucinols using bridgehead lithiation: The total synthesis of racemic clusianone and a formal synthesis of racemic garsubellin A

Article abstract of DOI:10.1021/jo070388h

(Chemical Equation Presented) The synthesis of polyprenylated phloroglucinol natural products, including clusianone, nemorosone, and garsubellin A, was pursued by a strategy involving construction of a core bicyclo[3.3.1]nonanetrione structure and subsequent elaboration via organolithium intermediates. Appropriate bridged core structures were obtained through the cyclization of a suitably substituted cyclohexanone enol ether or enol silane with malonyl dichloride. Additional substituents were then introduced by means of regioselective lithiation reactions, including the generation of bridgehead enolates, thus enabling the total synthesis of clusianone and also of an advanced intermediate toward nemorosone. In the case of garsubellin A, an additional THF-like ring was elaborated by a biomimetic 5-exo-tet cyclization of an enol ether (or enol) with a side-chain epoxide. This enabled a formal synthesis of racemic garsubellin A by accessing one of the late intermediates in the Danishefsky synthesis.

Full text of DOI:10.1021/jo070388h

Synthesis of Polyprenylated Acylphloroglucinols Using Bridgehead
Lithiation: The Total Synthesis of Racemic Clusianone and a
Formal Synthesis of Racemic Garsubellin A
Nadia M. Ahmad,† Vincent Rodeschini,† Nigel S. Simpkins,*,† Simon E. Ward,‡ and
Alexander J. Blake†
School of Chemistry, The UniVersity of Nottingham, UniVersity Park, Nottingham NG7 2RD, U.K., and
Medicinal Chemistry, Psychiatry Centre of Excellence for Drug DiscoVery, GlaxoSmithKline, New
Frontiers Science Park, Third AVenue, Harlow Essex CM19 5AW U.K.
nigel.simpkins@nottingham.ac.uk
ReceiVed March 5, 2007
The synthesis of polyprenylated phloroglucinol natural products, including clusianone, nemorosone, and
garsubellin A, was pursued by a strategy involving construction of a core bicyclo[3.3.1]nonanetrione
structure and subsequent elaboration via organolithium intermediates. Appropriate bridged core structures
were obtained through the cyclization of a suitably substituted cyclohexanone enol ether or enol silane
with malonyl dichloride. Additional substituents were then introduced by means of regioselective lithiation
reactions, including the generation of bridgehead enolates, thus enabling the total synthesis of clusianone
and also of an advanced intermediate toward nemorosone. In the case of garsubellin A, an additional
THF-like ring was elaborated by a biomimetic 5-exo-tet cyclization of an enol ether (or enol) with a
side-chain epoxide. This enabled a formal synthesis of racemic garsubellin A by accessing one of the
late intermediates in the Danishefsky synthesis.
Introduction
the core bicyclic structure, Figure 2.2 Type A structures,
including garsubellin, hyperforin, and nemorosone, have the acyl
residue at the C-1 bridgehead position, whereas type B systems
like clusianone have a C-3 acyl group. The more scarce type C
structures bear an acyl group at the bridgehead carbon C-5.
Additional elaboration of these structures, involving cyclization
of the olefinic (most frequently prenyl) side chains and the enolic
â-diketone, is commonplace, as exemplified by the furanoid ring
present in garsubellin A.
Perhaps the most well-known member of the PPAP family,
hyperforin (2), is thought to be the major bioactive constituent
from Hypericum perforatum (St. John’s wort, SJW), which has
well-known antidepressant properties.3 The use of SJW extracts
for the treatment of mild to moderate depression is widespread
Plants and trees of the family Clusiaceae (Guttiferae) are a
rich source of polyprenylated acylphloroglucinols (PPAPs).
These are characterized by a bicyclo[3.3.1]nonanetrione core
structure bearing additional acyl and prenyl substituents, e.g.,
garsubellin A (1), hyperforin (2), clusianone (3), and nemo-
rosone (4) (Figure 1).1
The various PPAPs have been divided into three classes,
depending on the relative position of the acyl substituent on
* Corresponding author. Tel: +44 (0)115 951 3533. Fax: +44 (0)115 951
3564.
† The University of Nottingham.
‡ GSK R&D.
(1) For a recent and comprehensive review of the area, see: (a) Ciochina,
R.; Grossman, R. B. Chem. ReV. 2006, 106, 3963-3986. See also: (b)
Verotta, L. Phytochem. ReV. 2002, 1, 389-407.
(2) Cuesta-Rubio, O.; Velez-Castro, H.; Frontana-Uribe, B. A.; Cardenas,
J. Phytochemistry 2001, 57, 279.
10.1021/jo070388h CCC: $37.00 © 2007 American Chemical Society
Published on Web 05/26/2007
J. Org. Chem. 2007, 72, 4803-4815
4803

Ahmad et al.
SCHEME 1. Stoltz’s Synthesis of the [3.3.1]Trione Core of 1
Our interest in this field first arose from the observation that
most PPAPs possess two substituted bridgehead positions
flanked by activating ketone functions. Having recently shown
that bridgehead lithiation-substitution processes in related
[3.3.1] systems were possible,11 we identified an opportunity
to apply this approach to the synthesis of PPAP natural products
and their analogues. The idea of accessing various types of
PPAPs by the late-stage affixing of appropriate substituents onto
a common [3.3.1] trione core system is an attractive one. In
particular, this strategy holds the promise of a very flexible
access to various types of natural PPAPs from a common core
structure, through appropriate functionalization of the C-1, C-3,
and C-5 positions. In addition, this would allow us to access
diverse unnatural analogue structures in order to probe the
largely unexplored SAR in these systems. This strategy,
however, required a rapid access to a suitable “core” trione
system. A solution to this problem was first proposed by
Spessard and Stoltz during their elegant approach to garsubellin
A (Scheme 1).12 In a modification of a procedure originally
described by Effenburger and co-workers,13 they were able to
convert the substituted enol silane 5 directly into the bridged
trione 6 by reaction with malonyl dichloride, with complete
diastereoselectivity at C-7.
FIGURE 1. Structure of some PPAPs.
FIGURE 2. Three types of PPAPs.
in Europe but remains controversial in terms of efficacy,
especially with regard to undesirable prescription drug interac-
tions.4 More recently isolated in 1997 from Garcinia subellip-
tica, garsubellin A (1) has since attracted significant synthetic
attention, in part due to its ability to induce choline acetyltrans-
ferase, a key enzyme involved in acetylcholine biosynthesis.5
Clusianone (3) has been known since 1976, and recent screening
has ascribed potent anti-HIV and cancer chemopreventive
properties to 3 and related compounds such as the regioisomeric
PPAP nemorosone (4).6,8a Although the structure of 3 was
originally determined by X-ray crystallography,7 more recently
there was some confusion of this compound with a C-7 epimer.8
The yield of bridged triketone product was modest, but the
ketone corresponding to the starting enol silane could also be
recovered. Unfortunately, enol derivatives having additional R-
and R′-substituents, destined to emerge as the C-1 and C-5
bridgehead groups, were even less satisfactory participants in
the cyclization. This deficiency provided an obvious spur for
our bridgehead substitution approach.
Despite the relatively low efficiency of the key Effenburger
cyclization, the brevity of this approach encouraged us to adopt
it for accessing a core structure, which could then be further
substituted by sequential regioselective lithiations. We envisaged
that target PPAPs 7 would be available by controlled substitution
at C-1, C-3, and C-5 of a vinylogous ester (O-alkylated 1,3-
diketone) such as 8 or its regioisomer. Compounds like 8 would
be prepared by Effenburger-type cyclization of appropriate
cyclohexanone-derived enol ethers or enol silanes 9, themselves
available by fairly routine and concise sequences starting from
a suitable cyclohexan-1,3-dione derivative 10 (Scheme 2).
We reasoned that regioselectivity for metalation at one
bridgehead position should be possible as a result of the different
The broad range of biological activities exhibited by this
family of compounds, together with the challenging structure
of PPAPs, has stimulated much synthetic activity, and several
groups have described progress toward specific members of the
PPAP family.1a Only very recently, however, has 1 succumbed
to total synthesis by the groups of Shibasaki9 and Danishefsky.10
(3) (a) Structure: Bystrov, N. S.; Chernov, B. K.; Dobrynin, V. N.;
Kolosov, M. N. Tetrahedron Lett. 1975, 2791-2794. (b) Biosynthesis:
Adam, P.; Arigoni, D.; Bacher, A.; Eisenreich, W. J. Med. Chem. 2002,
45, 4786-4793.
(4) Madabushi, R.; Frank, B.; Drewelow, B.; Hartmut, D.; Butterweck,
V. Eur. J. Clin. Pharm. 2006, 62, 225-233.
(5) Fukuyama, Y.; Kuwayama, A.; Minami, H. Chem. Pharm. Bull. 1997,
45, 947-949.
(6) (a) Ito, C.; Itoigawa, M.; Miyamoto, Y.; Onoda, S.; Sundar, Rao,
K.; Mukainaka, T.; Tokuda, H.; Nishino, H.; Furukawa, H. J. Nat. Prod.
2003, 66, 206-209. (b) US Patent US 2005/0090693 A1.
(7) McCandlish, L. E.; Hanson, J. C.; Stout, G. H. Acta Crystallogr.,
Sect B 1976, 32, 1793-1801.
(8) (a) Piccinelli, A. L.; Cuesta-Rubio, O.; Chica, M. B.; Mahmood, N.;
Pagano, B.; Pavone, M.; Barone, V.; Rastrelli, L. Tetrahedron 2005, 61,
8206-8211. (b) Delle Monache, F.; Delle, Monache, G.; Gacs-Baits, E.
Phytochemistry 1991, 30, 2003-2005.
(9) Kuramochi, A.; Usuda, H.; Yamatsugu, K.; Kanai, M.; Shibasaki,
M. J. Am. Chem. Soc. 2005, 127, 14200-14201.
(10) Siegal, D. R.; Danishefsky, S. J. J. Am. Chem. Soc. 2006, 128,
1048-1049.
(11) Giblin, G. M. P.; Kirk, D. T.; Mitchell, L.; Simpkins, N. S. Org.
Lett. 2003, 5, 1673-1675.
(12) Spessard, S. J.; Stoltz, B. M. Org. Lett. 2002, 4, 1943-1946.
(13) Scho¨nwa¨lder, K-H.; Kollatt, P.; Stezowski, J. J.; Effenburger, F.
Chem. Ber. 1984, 117, 3280-3296.
4804 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
SCHEME 2. A General Strategy To Access Various PPAPs
SCHEME 3. Synthesis of Enol Derivatives for Effenberger Cyclization
steric environments for positions C-1 and C-5 resulting from
the gem-dimethyl substituents at C-8. The position of diketone
alkylation in structures like 6 might also be expected to affect
the regioselectivity of bridgehead metalation, and there was
some precedent from the Stoltz work that single regioisomeric
vinylogous ester structures could be prepared (vide infra).
The introduction of substituents at the vinylic position (C-3)
by direct metalation also appeared to be viable according to
literature precedents.14 Careful planning in the order of introduc-
tion of bridgehead vs C-3 prenyl groups would be required in
cases where regioselective side chain oxidation-cyclization is
required (e.g., garsubellin A). For our initial targets, we also
wanted to avoid oxidation of the exo-orientated C-7 prenyl
substituent, and it was unclear if we could rely on steric effects
or directing effects (e.g., from the enolized diketone system) in
this regard. A tactic involving the use of a C-7 allyl group as a
less reactive “bystander” group capable of ultimately being
converted into a prenyl group looked attractive.
In this full report, we present the full details of this synthesis,
along with our synthetic efforts toward nemorosone (4), and
we present a formal synthesis of garsubellin A (1) by a new
access to an advanced intermediate previously prepared by the
Danishefsky group.10
Results and Discussion
1. Synthesis of Clusianone. As installation of a substituent
at the hindered C-1 position of 8 was considered an extreme
test of our strategy (vide infra), we first decided to focus on
substitution reactions at C-3 and C-5 of the core trione
derivative. The synthesis started with known vinylogous ester
11,16 bearing a prenyl group at the position which would become
C-1 in the cyclized product (Scheme 3).
Prenylation of 11 using LDA and prenyl bromide gave 12,
which was then reacted with MeLi and hydrolyzed under mildly
SCHEME 4. Effenberger Cyclization
In our previous communication, we described a concise
synthesis of clusianone (3) using the approach outlined above.15
(14) (a) Miyata, O.; Schmidt, R. R. Tetrahedron Lett. 1982, 23, 1793-
1796. (b) For a more recent example, see: Zapf, C. W.; Harrison, B. A.;
Drahl, C.; Sorensen, E. J. Angew. Chem., Int. Ed. 2005, 44, 6533-6537.
(15) Rodeschini, V.; Ahmad, N. M.; Simpkins, N. S. Org. Lett. 2006, 8,
5283-5285.
J. Org. Chem, Vol. 72, No. 13, 2007 4805

Ahmad et al.
SCHEME 5. Synthesis of Bridged Trione and Trione-Ether Derivatives
acidic conditions to give enone 13. Conjugate addition of a
SCHEME 6. Proposed Mechanism for the Reaction of an
Enol Ether with Malonyl Dichloride
methyl group to this tetrasubstituted system proved more
difficult than anticipated. The use of stoichiometric higher order
cuprate reagents (Me2CuLi), alone or in the presence of Lewis
acids (TMSCl, BF3-OEt2), failed to deliver useful quantities
of ketones 14/15. In contrast, copper-catalyzed Grignard addition
in the presence of HMPA and TMSCl provided the ketone
product as a mixture of diastereoisomers in a satisfying 88%
yield. The cis derivative 14 was the major component, as
determined by NOE experiments. This was of no consequence,
however, as the mixture of diastereoisomers was easily con-
verted into a regioisomeric mixture of enol ethers 16 and 17 in
a combined 87% yield under thermodynamic conditions.17
Alternatively, enol silane 18 was obtained as a single isomer
starting from the cis-diprenylated ketone 14 under kinetic
conditions in 82% yield.
Having obtained a few appropriate enol ether and enol silane
substrates, we started to explore the Effenburger-type cycliza-
tion. Originally, the Effenberger group reported quite efficient
cyclizations, but these were based on the use of the enol
component in excess. For example, reaction of 4 equiv of methyl
enol ether 19 with 1 equiv of malonyl dichloride, at -19 °C,
using Et2O as solvent, allowed the isolation of the cyclized
product 20 in 84% yield (Scheme 4).
However, the use of more elaborate enol ethers, such as 16-
18, in excess was not attractive in our case, and we preferred
to start our study with the modified conditions described by
Stoltz, which employed malonyl dichloride in excess.
Under these conditions, reaction of enol ether mixture 16/17
with malonyl dichloride (2 equiv) in Et2O at -20 °C for 24 h,
followed by a basic workup, furnished the desired trione 21 as
a single diastereomer (ca. 30% crude material), which could be
separated by base extraction from recovered ketones 14/15 (55%
recovered, Scheme 5). In this case, both regioisomeric enol
methyl ethers appear to participate in the process. Essentially
the same results were achieved by using the TBS enol ether
18. A rapid screening of conditions indicated that either CH2-
Cl2 or Et2O could be used as the solvent (although in general
Et2O gave cleaner results) and that the number of equivalents
of malonyl dichloride could be reduced to 1 equiv without
reducing conversion or product yield.
Having viable quantities of triketone 21, we next explored
its conversion into the corresponding vinylogous ester as a
means to protect the 1,3-diketone function during the planned
lithiation chemistry. O-Methylation under basic conditions
provided an equimolar mixture of the regioisomeric vinylogous
esters 22 and 23, which could easily be separated by column
chromatography. In contrast, under acidic conditions (PTSA),
used by Stoltz to prepare a similar enol ether, completely
regioselective methylation occurred providing 23 as the sole
product in 24% overall yield from 16 and 17 (or alternatively
from 18). The contrasting results were shown to be due to the
greater thermodynamic stability of regioisomer 23, as demon-
strated by converting 22 completely into 23 (no 22 remained at
the end of the reaction) under the same acidic conditions
(Scheme 5).
Although this short sequence of reactions gave us rapid access
to our desired core [3.3.1] trione structure, the consistently low
conversions obtained in the key cyclization were frustrating,
and we decided to investigate alternative conditions for this
reaction. The mechanism for this transformation, as originally
proposed by Effenburger et al., involved the initial conversion
of malonyl dichloride into ketene B, the intervention of one
(16) Majetich, G.; Hull, K.; Casares, A. M.; Khetani, V. J. Org. Chem.
1991, 56, 3958-3973.
(17) Heiszwolf, G.J.; Kloosterziel, H. J. Chem. Soc., Chem. Commun.
1966, 51.
4806 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
SCHEME 7. Completion of the Synthesis of Clusianone
molecule of enol ether serving to mop up HCl and forming
chloroether A (Scheme 6). Electrophilic ketene B then reacts
with the remaining enol ether diastereoselectively through an
axial approach, delivering the malonyl unit on the opposite face
to the substituent on position that emerges as C-7. Proton
exchange on oxonium species C generates a transposed enol
ether D, which finally undergoes a Dieckmann type reaction
leading to the observed cyclized product.
This mechanism might account for the good yields obtained
by the original authors when using the enol ether component
in excess. Since we required use of the valuable enol ether (or
enol silane) as the limiting component in our chemistry we
sought to form the putative ketene by the more conventional
action of a base. Following this idea, we tried to generate the
ketene in situ by adding a bulky amine to the reaction mixture.
However, in the presence of Huenig’s base (iPr2NEt) or proton
sponge (1,8-bis(dimethylamino)naphthalene), no cyclized prod-
uct was obtained. The mechanism proposed above also requires
the regeneration of an enol ether from the oxonium species B,
and such a process has been shown to be particularly favored
in the case of triisopropylsilyl enol ether derivatives.18 In our
case, however, the use of TIPS enol ether 2419 led to no
cyclization at all, and 24 was recovered unaffected after the
workup.
cyclization using bis(cyclopentadienyl)hafnium dichloride as a
Lewis acid mediator, but we were unable to improve the yields
of bicyclic adducts using this approach.21 After expending some
considerable effort, we found ourselves unable to improve on
the modest cyclization yields reported above and decided to
proceed with the more advanced aspects of the synthesis.
Although the key cyclization reaction was (predictably) low
yielding, this short sequence provided viable quantities of the
cyclized products, and the recovered ketones 14/15 could easily
be recycled through conversion to 16/17 to provide more of
the cyclized product. We then established that either regioiso-
meric ether 22 or 23 could be converted into the natural product
clusianone, as shown in Scheme 7.
The total synthesis was first accomplished starting from
isomer 23, which, on treatment with excess LDA, followed by
alkylation with prenyl bromide, gave 26 in an excellent 91%
yield. For this transformation, the use of 2 equiv of base proved
optimal, although no evidence could be found for dianion
formation (e.g., by quenching with deuterium sources). When
applied to isomer 22, the same conditions gave the desired
product 28, although in a lower 67% yield. It therefore appears
that the orientation of the vinylogous ester is of little significance
in this particular metalation-substitution process, which is
perhaps controlled mainly by the ketone in the one-carbon
bridge.
Acylation at C-3 was accomplished by the action of LTMP
followed by quenching the intermediate vinyllithium with
benzoyl chloride. The process was equally efficient for both
regioisomers 26 and 28, providing the fully substituted deriva-
tives 27 and 29, respectively. Finally, hydrolysis of the
vinylogous ester 27, under conditions akin to those described
previously,12 gave (()-clusianone in 90% yield. Hydrolysis of
regioisomer 29 proved much more difficult, and under the same
conditions (LiOH, dioxane, 90 °C), no deprotection occurred
The use of an alternative electrophile, the so-called “magic
malonate” 25,20 similarly led to no annulation under otherwise
similar conditions. Spessard and Stoltz had described one
(21) During the preparation of this manuscript, another synthesis of
clusianone appeared, using a similar approach, where improved yields of
bridged adduct were obtained from the Effenberger cyclization by use of
BF3‚OEt2 as Lewis acid mediator; see: Nuhant, P.; David, M.; Pouplin,
T.; Delpech, B.; Marazano, C. Org. Lett. 2007, 9, 287-289.
(18) Magnus, P.; Mugrage, B. J. Am. Chem. Soc. 1990, 112, 462-464.
(19) Derivative 18 was prepared in the same way as for TBS enol ether
14; see the Supporting Information for details.
(20) Kappe, T. Monatsch. Chem. 1967, 98, 883-895.
J. Org. Chem, Vol. 72, No. 13, 2007 4807

Ahmad et al.
SCHEME 9. Preparation of Bicyclo[3.3.1]nonane
SCHEME 8. Synthesis of Enol Silanes 33 and 34
Derivatives 39 and 40
the neighboring gem-dimethyl groups in the diketones 31 and
32. This would be expected to result in noncoplanar carbonyl
functions, thereby reducing the acidity of the methine proton
at the normally “doubly activated” position (labeled C-1), which
is also in a very hindered position.
even after prolonged reaction times or at a higher temperature.
In contrast, when submitted to Krapcho-type conditions (LiCl,
DMSO, 120 °C),22 29 was readily deprotected to give 3 in 57%
yield. Clusianone was thus obtained as a mixture of enol
tautomers, with spectroscopic data identical to those reported
for the natural product.8a Further evidence for the identity of
our synthetic sample was obtained by comparison of the spectral
data of intermediate 27 with those of a previously reported
methylated derivative of clusianone.23 Despite the low yields
for the key cyclization, the synthesis of 3 was concisely achieved
in only nine synthetic steps with a 12% overall yield.
2. Progress toward a Synthesis of Nemorosone. Encouraged
by the brevity and overall efficiency of our route to clusianone,
we then decided to further probe our strategy by applying it to
the synthesis of the close isomer of clusianone, nemorosone
(4). Adoption of exactly the same strategy as beforesnamely
pre-installation of the eventual C-1 bridgehead substituent in
the enol silaneswould require an electronically very different
cyclization partner for the Effenburger procedure. In this case,
the benzoyl substituent required in the bridged product neces-
sitates the use of an enol ether from a 1,3-diketone. Hoping
that this dramatic change might influence the yields of the
reaction favorably, we proceeded to prepare a suitably substi-
tuted enol silane bearing a benzoyl group at position C-1
(nemorosone numbering) (Scheme 8).
Upon attempted reaction with malonyl dichloride under our
previously determined conditions, these derivatives failed to give
any cyclized product. Following this fruitless diversion, we
decided to return to the use of enol silane precursors in which
one bridgehead prenyl group was “pre-installed”, with the
intention of introducing the bridgehead benzoyl substituent at
C-1 late on. Following this strategy, we next prepared the bis-
prenylated cyclohexanone 36 (Scheme 9).
The revised sequence started with the prenylation of enone
30, followed by copper catalyzed methyl conjugate addition to
give 36 (9:1 trans/cis),24 which was then straightforwardly
converted into enol silane 37. As observed before, the cyclization
produced a moderate yield of the trione 38, together with some
recovered ketone 36. Crude trione 38 was directly protected as
a mixture of regioisomers 39 and 40, with an overall 29% yield
starting from enol silane 37, the regio and stereochemistry of
the major isomer 39 being confirmed by X-ray crystallography
(see the Supporting Information).
With both regioisomers in hand, the next task was the
installation of the required prenyl substituent at the C-3 vinylic
position. This proved much more sluggish than the acylation
carried out previously in the case of clusianone synthesis. After
deprotonation of 39 or 40 using LTMP, no substantial reaction
with an excess of prenyl or allyl bromide occurred at -78 °C,
only starting material being recovered. While changing the base
to LDA or warming the reaction mixture to 0 °C did not improve
the outcome, we were pleased to find that the use of a mixed
higher order cuprate nicely solved this issue. Thus, after
deprotonation with LTMP, the resulting vinyllithium species
was transmetalated at -40 °C using freshly prepared Lipschutz’s
Conjugate methyl addition to the known enone 30,12 followed
by quenching of the intermediate enolate with benzaldehyde,
afforded a mixture of aldol products which, following Dess-
Martin periodinane oxidation, gave the easily separated isomeric
â-diketones 31 and 32. To our initial surprise, reaction of each
diketone 31 and 32 with TBSCl under thermodynamic condi-
tions led to the formation of a nonconjugated enol silane
derivatives 33 and 34, respectively. This is presumably a result
of the steric interaction between the benzoyl substituent and
(22) Krapcho, A. P.; Weimaster, J. F.; Eldridge, J. M.; Jahngen, E. G.
E., Jr.; Lovey, A. J.; Stephens, W. P. J. Org. Chem. 1978, 43, 138.
(23) De Oliveira, C. M. A.; Porto, A. M.; Bittrich, V.; Vencato, I.;
Marsaioli, A. J. Tetrahedron Lett. 1996, 37, 6427-6430.
(24) The trans stereochemistry of derivative 35, and thus of 36, was
tentatively assigned as a result of the favored axial approach of prenyl
bromide toward the transient enolate species derived from 30.
4808 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
SCHEME 10. Copper-Mediated Vinylic Alkylation and Attempted Bridgehead Substitution
SCHEME 11. Formation of Ether 45
SCHEME 12. Synthesis of Iodo Derivative 47
higher order cyanocuprate.25 The resulting mixed cyanocuprate
then reacted smoothly with prenyl bromide, and thus, derivatives
41 and 42 were obtained in good yields (Scheme 10). Having
obtained these two polyprenylated derivatives, all that was
needed to achieve the synthesis of 4 was to introduce a benzoyl
group at bridgehead position C-1 and to hydrolyze the vinylo-
gous ester.
failed and bridgehead substitution did not occur. Instead, the
C-3 prenyl residue underwent a formal dehydrogenation to give
conjugated diene 43 in 60% yield (Scheme 10). A similar
observation was made using regioisomeric dione 41 as the
starting material, but the transformation was much lower-
yielding and the product difficult to separate from other by-
products.
Disappointingly, our initial misgivings concerning the vi-
ability of bridgehead substitution proximal to the gem-dimethyl
groups were now shown to have good foundation, and all our
attempts to substitute at C-1 of 41 or 42 using LDA or LTMP
in conjunction with various electrophiles failed. This result was
frustrating, taking into account the contemporaneous report of
Siegel and Danishefsky of bridgehead iodination at this position
on a similar (but more complex) substrate, albeit in low yield,
during a synthesis of garsubellin A (vide infra). Their sequence
used initial iodination using LDA and TMSCl, followed by an
iodine-magnesium exchange and electrophilic quenching (vide
infra). The metalation conditions appeared substantially different
to our standard procedures, and required use of excess LDA
with TMSCl as the in situ quench, initially for 2 min at -78 °C
and then 12 min at ice bath temperature, prior to iodine
quenching. When applied to our substrate 42, these conditions
We decided to explore substitution at the troublesome C-1
position in more detail and prepared derivative 44 from 40 using
LTMP and benzoyl chloride, where the seemingly sensitive
prenyl residue at C-3 was replaced by a benzoyl moiety. In this
case, bridgehead prenylation would have provided us with an
alternative route to clusianone. Quite surprisingly, upon treat-
ment with LTMP at low temperature, no sign of bridgehead
substitution was observed, but instead ether 45, arising via
deprotonation of the methyl ether, was isolated in 43% yield
(Scheme 11).
Eventually, we found that using the C-3 trimethylsilyl
derivative 46 in conjunction with Danishefsky’s conditions
allowed isolation of bridgehead iodide 47 in 40% yield
(Scheme 12).10 This result is significant in providing an
additional example of metalation at the sterically congested C-1
position (vide infra), but the modest yields in this step and the
earlier cyclization make the approach unattractive.
(25) Lipshutz, B. H.; Koerner, M.; Parker, D. A. Tetrahedron Lett. 1987,
28, 945-948.
J. Org. Chem, Vol. 72, No. 13, 2007 4809

Ahmad et al.
SCHEME 13. Danishefsky’s Final Steps toward Garsubellin A
SCHEME 14. Synthesis of Bicyclo[3.3.1]nonane Derivative
55
verted to the more stable ether 55 with complete regioselectivity
for the formation of enol at C-4 by heating to reflux in acidic
MeOH.
The incorporation of an allyl substituent at C-7 rather than a
prenyl appendage, in this series of compounds was dictated by
our anticipation of bridgehead prenylation at C-5 of 55, followed
by side chain epoxidation. The higher reactivity of a trisubsti-
tuted alkene over a terminal one in reactions with peracids
andrelated oxidants was expected to secure the desired regi-
oselectivity in this latter step. The synthesis proceeded, as shown
in Scheme 15, culminating in the synthesis of the Danishefsky
intermediate 48.
Bridgehead prenylation of 55 proved more problematic than
in our clusianone work, and gave the desired compound 56 in
only 46% yield. Since up to 91% yield had been obtained in
the analogous prenylation of the methyl ether 23, we were
initially tempted to assign the difficulties to the presence in 55
of two free bridgehead positions and a free C-3 position.
However, bearing in mind the extreme difficulties experienced
in C-1 substitution in other systems, and the sluggish prenylation
seen at C-3 in the absence of copper, we considered competing
substitution at these positions unlikely. Indeed, we were unable
to isolate products from reaction at either the C-1 bridgehead
position or the C-3 vinylic position. Both regio- and stereo-
chemistry of derivative 56 were secured by X-ray crystal-
lography (see the Supporting Information).
It appears that bridgehead substitution at C-1 is much less
facile than that at C-5 and that metalation at other positions on
this type of substrate can often compete successfully. Although
the preparation of iodide 47 provides some encouragement that
a synthesis of nemorosone might be possible by this route, we
instead turned our attention to the synthesis of the more elaborate
PPAP derivative, garsubellin A.
3. Formal Synthesis of Garsubellin A. It was during the
course of our studies that the aforementioned Siegel and
Danishefsky synthesis of garsubellin A appeared, which featured
i
a bridgehead metalation for the installation of the C-1 PrCO
acyl group. During the final steps of their sequence, they were
able to convert the allyl groups in intermediate 48 into the
required prenyl groups of 49, using Grubbs’ second-generation
metathesis catalyst. This was followed by the bridgehead
iodination, iodine-magnesium exchange, and electrophilic
quenching with isobutyraldehyde to give 50. Subsequent oxida-
tion and deprotection of the tertiary alcohol then led to
garsubellin A (Scheme 13).
We thought that the use of the Effenburger cyclization would
give us rapid access to the advanced tricyclic intermediate 48
and thus could provide an alternative route to garsubellin A.
The synthesis would also provide further examples of selective
bridgehead substitution, as well as providing new chemistry
for the additional THF-type of ring present in this natural
product.
We commenced our synthesis with the known enone 51,26
which was converted into ketone 52 by the copper-catalyzed
Michael addition of MeMgBr (Scheme 14). Cyclohexanone 52
was easily converted into the corresponding enol silane 53
(the structure shown was the major component in a ca. 5:1
regioisomeric mixture), which was then subjected to our
modified Effenburger cyclization protocol. This process fur-
nished the desired bicyclic system 54 in a modest but usable
35-40% yield, and the crude product was immediately con-
Stereocontrol in the oxidation of the bridgehead prenyl group,
required for the formation of the THF ring of 1, has not been
fully addressed. For example, Shibasaki’s dihydroxylation
approach was totally non-diastereoselective. We chose to focus
on a (probably) biomimetic side-chain epoxidation-epoxide
opening strategy but found a similar lack of diastereoselectivity.
Epoxidation of 56 using m-CPBA did not give clean conver-
sions, and the use of DMDO gave a rather unstable mixture of
diastereomeric epoxides. Eventually, we found that combining
DMDO epoxidation with a novel THF-forming cyclization
mediated by TMSCl enabled the efficient formation of isomers
57 and 58 as a 1:1 mixture. Trimethylsilyl chloride was found
to be much more effective than the corresponding iodide for
this cyclization, which we assume proceeds via silicon-mediated
activation of the epoxide. Surprisingly, however, we have not
been able to directly isolate the silicon-protected alcohol
derivatives from this reaction, but these compounds (59 and
60) could be prepared by silylation under standard conditions.
At this stage, the stereochemical assignment of these two
isomers was tentative, based on comparison of our NMR data
with that in the literature, the assignment being demonstrated
as correct by subsequent transformations.
The correct side chain isomer 59 was then subjected to C-3
lithiation and allylation with recourse to the mixed higher
order cuprate as described previously. Using this protocol we
were able to access the Danishefsky intermediate in protect-
ed form (61), and subsequent treatment with Et3N-HF then
(26) Johnson, W. S.; McCarry, B. E.; Markezich, R. L.; Boots, S. G. J.
Am. Chem. Soc. 1980, 102, 352-359.
4810 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
SCHEME 15. Alternative Access to Intermediate 48
SCHEME 16. Modified Approach to THF Installation
provided 48 identical in all respects with that reported previously
formation of relatively elaborate tricyclic structures (57/58) from
(Scheme 15).
a simple, readily available enol silane precursor.
The route described is quite a step-efficient approach to
intermediate 48; however, the poor yields in some of the key
steps and the lack of stereoselectivity in the side-chain oxidation
detract significantly. In an attempt to address these issues, we
briefly examined a variant of the above route in which the C-5
prenyl group was installed prior to the Effenburger cyclization
(Scheme 16).
Conclusion
The present study serves to further delineate the potential
for an Effenburger-type of cyclization in the synthesis of PPAP
natural products and their analogues while highlighting the
shortcomings of the currently available protocols. By teaming
this powerful transformation with regioselective lithiation
processes, we have been able to achieve the synthesis of
clusianone (3) in nine steps, and Danishefsky’s intermediate
48 for garsubellin A (1) has been obtained in a 10-step sequence
for our shortest route, although with modest yields. We have
also shown that the THF ring of garsubellin A can be obtained
through a biomimetic epoxidation-cyclization sequence, with
modest selectivity. Our strategy holds the promise of flexibility
in the installation of different substituents at C-1, C-3, and C-5,
which is an important feature for probing the intriguing
biological activities of these systems. A limitation is the
difficulty in introducing substituents at the sterically congested
C-1 bridgehead position, which at the moment prevents easy
access to nemorosone. Work within our group is ongoing to
supersede the existing method with a more efficient variant. In
addition, we are accessing nonracemic PPAPs by employing
kinetic resolution in the bridgehead lithiation by use of chiral
base reagents, and we are currently probing the SAR for this
series of PPAPs by screening our final products and intermedi-
ates for their biological activities.
Enol silane 64 was prepared in a similar way to the
bis-prenylated analogue 37. Prenylation of enone 51 afforded
62 and was followed by copper-catalyzed Grignard addition to
give ketone 63 (as a mixture of trans and cis isomers), which
was then efficiently converted into the more substituted enol
silane 64 in 95% yield. Then, cyclization using malonyl
dichloride provided ∼30% of crude 65 (together with 58%
ofrecovered ketone 63). The synthesis of 65 provided an
opportunity to re-assess side chain oxidation using a compound
with an unprotected 1,3-diketone, and rather than purifying the
cyclized product at this stage (which resulted in mass losses),
we found it more efficient to react it directly with m-CPBA.
Pleasingly, this epoxidation proved effective and led directly
to the previously synthesized alcohols but now with a diaster-
eomeric ratio favoring the desired isomer 57.
This revised approach has the advantage of avoiding the
modest yielding bridgehead prenylation step as well as the
formation of a vinylogous ester (cf. 55), and the selectivity in
the epoxidation is also improved. Balanced against this is the
drop in yield for the Effenburger cyclization. Nevertheless, the
overall process is quite step-efficient, allowing the rapid
J. Org. Chem, Vol. 72, No. 13, 2007 4811

Ahmad et al.
acidified to pH ∼1 using 2 M HCl, and then extracted using DCM
(4 × 25 mL). The combined organic layers were dried over MgSO4,
and the solvent was removed in vacuo to provide crude 21 as an
orange foam (190 mg): 1H NMR (400 MHz, CDCl3) 5.96 (s, 1H),
4.98 (m, 1H), 4.71 (m, 1H), 3.23 (m, 1H), 2.53 (m, 2H), 2.10 (m,
2H), 1.77-1.52 (m, 4H), 1.65 (s, 3H), 1.64 (s, 3H), 1.56 (s, 3H),
1.55 (s, 3H), 1.10 (s, 3H), 0.79 (s, 3H); 13C NMR (100 MHz,
CDCl3) 208.2, 192.5, 184.7, 133.8, 133.3, 122.5, 119.8, 109.3, 68.1,
56.2, 45.8, 41.1, 32.9, 28.2, 25.9, 25.8, 23.8, 23.2, 18.1, 17.9, 16.0;
HRMS (ESI) m/z calcd for C21H31O3 331.2268, found 331.2252,
[M + H]+. Crude 21 obtained above was dissolved in acetone (15
mL), K2CO3 (320 mg, 4 equiv) and Me2SO4 (73 µL, 1.2 equiv)
were added, and the resulting suspension was heated under reflux
for 2 h under N2. The mixture was cooled to rt, the solid was
removed by filtration, and the solvent was removed in vacuo. The
residue was purified by column chromatography (eluent petroleum
ether/CH2Cl2/AcOEt 70/28/2, then 50/40/10) to give pure 23 (73
mg) and then pure 22 (71 mg) in 25% combined yield over two
steps. 22: Rf ) 0.27 (petroleum ether/AcOEt 4/1); white amorphous
solid; FTIR (CHCl3) 2930, 1730, 1650, 1590, 1347 cm-1; 1H NMR
(500 MHz, CDCl3) 5.82 (s, 1H), 4.96 (m, 1H), 4.73 (m, 1H), 3.72
(s, 3H), 3.22 (dd, J ) 4.5, 2.5, 1H), 2.58 (br dd, J ) 14.0, 5.0,
1H), 2.51 (dd, J ) 14.0, 8.0, 1H), 2.13 (ddd, J ) 13.5, 4.0, 3.0,
1H), 2.08 (m, 1H), 1.75-1.56 (m, 3H), 1.66 (s, 3H), 1.65 (s, 3H),
1.60 (s, 3H), 1.54 (s, 3H), 1.08 (s, 3H), 0.78 (s, 3H); 13C NMR
(100 MHz, CDCl3) 207.4, 195.2, 177.6, 133.5, 133.3, 122.5, 119.7,
107.7, 64.4, 61.7, 56.3, 45.2, 41.3, 34.2, 28.6, 25.9, 25.8, 23.9, 23.8,
18.0, 17.9, 16.1; HRMS (ESI) m/z calcd for C22H32O3Na 367.2244,
found 367.2245. 23: Rf ) 0.31 (petroleum ether/AcOEt 4/1);
Experimental Section
(()-(4S,6S)-1-tert-Butyldimethylsilyloxy-5,5-dimethyl-4,6-bis-
(prenyl)cyclohex-1-ene (18). A solution of BuLi (1.42 mL, 1.6
n
M in hexanes, 2.27 mmol, 1.2 equiv) was added dropwise to a
solution of iPr2NH (345 µL, 2.46 mmol, 1.3 equiv) in THF (3 mL)
at 0 °C. After 15 min, the LDA solution was cooled to -78 °C,
and a solution of ketone 14 (495 mg, 1.89 mmol) in THF (3 mL)
was transferred via cannula. The resulting mixture was stirred 30
min at -78 °C, and then TBSOTf (651 µL, 2.84 mmol, 1.5 equiv)
was added. The resulting mixture was allowed to warm slowly to
0 °C and then quenched with saturated aqueous NaHCO3. The layers
were separated, and the aqueous phase was extracted using
petroleum ether. The combined organic phases were dried over
MgSO4 and concentrated to a colorless oil, which was purified by
column chromatography (eluent petroleum ether/Et2O/Et3N 98/1/
1) to give pure 18 (582 mg, 82%): Rf ) 0.75 (petroleum ether/
Et2O 95/5); colorless oil; FTIR (CHCl3) 2950, 1664, 1385 cm-1
;
1H NMR (500 MHz, CDCl3) 5.25 (m, 1H), 5.09 (m, 1H), 4.75 (m,
1H), 2.24-2.14 (m, 3H), 2.04 (dtd, J ) 17.0, 5.0, 2.5, 1H), 1.91
(m, 1H), 1.79-1.61 (m, 2H), 1.69 (s, 3H), 1.65 (s, 3H), 1.60 (s,
6H), 1.23 (m, 1H), 0.98 (s, 3H), 0.90 (s, 9H), 0.73 (s, 3H), 0.12 (s,
6H); 13C NMR (100 MHz, CDCl3) 152.3, 131.6, 128.5, 127.5,
124.4, 102.0, 51.1, 44.9, 36.8, 28.2, 27.1, 26.9, 26.0, 25.9, 25.8,
25.7, 18.3, 17.9, 17.8, 16.2, -4.2, -4.3; HRMS (ESI) m/z calcd
for C24H45OSi 377.3234, found 377.3229 [M + H]+.
(()-(4S,6S)-(1-Triisopropylsilyloxy-5,5-dimethyl-4,6-bis(pre-
n
nyl)cyclohex-1-ene (24). A solution of BuLi (937 µL, 1.6 M in
hexanes, 1.5 mmol, 1.5 equiv) was added dropwise to a solution
i
of Pr2NH (210 µL, 1.5 mmol, 1.5 equiv) in THF (5 mL) at 0 °C.
colorless oil; FTIR (CHCl3) 2930, 1728, 1644, 1612, 1377 cm-1
;
After 15 min, the LDA solution was cooled to -78 °C, and a
solution of ketone 14 (266 mg, 1.01 mmol) in THF (3 mL) was
transferred via cannula. The resulting mixture was stirred for 15
min at -78 °C, and then TIPSOTf (463 µL, 1.70 mmol, 1.7 equiv)
was added. The resulting mixture was allowed to warm to 0 °C
and then quenched with saturated aqueous NaHCO3. The layers
were separated, and the aqueous phase was extracted using
petroleum ether. The combined organic phases were dried over
MgSO4 and concentrated to a colorless oil, which was purified by
column chromatography (eluent petroleum ether/Et3N 99/1) to give
pure 24 (360 mg, 85%): Rf ) 0.54 (petroleum ether); colorless
1H NMR (500 MHz, CDCl3) 5.73 (s, 1H), 4.95 (m, 1H), 4.66 (m,
1H), 3.75 (s, 3H), 3.18 (br s, 1H), 2.59 (dd, J ) 14.0, 6.5, 1H),
2.41 (dd, J ) 14.0, 5.8, 1H), 2.12 (m, 1H), 2.02 (m, 1H), 1.75-
1.65 (m, 3H), 1.67 (2 s, 6H), 1.58 (s, 3H), 1.55 (s, 3H), 1.08 (s,
3H), 0.77 (s, 3H); 13C NMR (125 MHz, CDCl3) 208.3, 196.7, 174.4,
133.5, 133.1, 122.6, 120.1, 106.6, 70.8, 56.7, 53.2, 46.3, 41.0, 32.7,
27.9, 25.9, 25.8, 24.1, 22.6, 18.1, 17.9, 16.0; HRMS (ESI) m/z calcd
for C22H33O3 345.2435, found 345.2422 [M + H]+.
Isomerization of 22 To Give 23. Ester 22 (150 mg, 435 µmol)
was dissolved in MeOH (5 mL), trimethyl orthoformate (1 mL)
and PTSA (10 mg) were added, and the resulting solution was
heated at 50 °C for 48 h under N2. The mixture was cooled to
room temperature, Et3N (2 drops) was added, and the mixture was
concentrated in vacuo. The residue was purified by column
chromatography (eluent petroleum ether/AcOEt 9/1 then 4/1) to
give pure 23 (97 mg, 65%).
(()-(1S,5R,7S)-2-Methoxy-8,8-dimethyl-1,5,7-tris(prenyl)-
bicyclo[3.3.1]non-2-ene-4,9-dione (28). At -78 °C, prenyl bromide
(176 µL, 1.53 mmol, 5 equiv) was added to a solution of 22
(105 mg, 305 µmol) in THF (12 mL), followed by a solution of
LDA (1.53 mL, 0.5 M, 2.5 equiv, prepared by adding a solution of
nBuLi (625 µL, 1 mmol, 1.6 M in hexanes) to a 0 °C solution of
iPr2NH (140 µL, 1 mmol) in THF (1.23 mL). The resulting solution
was stirred for 15 min at -78 °C and then quenched with saturated
aqueous NH4Cl and allowed to reach rt. The layers were separated,
and the aqueous phase was extracted using Et2O. The combined
organic phases were dried over MgSO4 and concentrated to an oil,
which was purified by column chromatography (eluent petroleum
ether/AcOEt 95/5 to 9/1) to give pure 28 (85 mg, 67%): Rf ) 0.21
(petroleum ether/AcOEt 9/1); colorless oil; FTIR (CHCl3) 2982,
1725, 1644, 1596, 1376, 1271 cm-1; 1H NMR (400 MHz, CDCl3)
5.81 (s, 1H), 4.98 (m, 2H), 4.72 (m, 1H), 3.69 (s, 3H), 2.60 (br d,
J ) 14.0, 1H), 2.50 (dd, J ) 14.0, 8.0, 1H), 2.43 (m, 2H), 2.06
(dm, J ) 13.5, 1H), 1.88 (dd, J ) 13.5, 4.5, 1H), 1.70 (m, 1H),
1.66 (s, 12H), 1.66 (m, 1H), 1.64 (s, 3H), 1.53 (s, 3H), 1.32 (t, J
) 13.5, 1H), 1.04 (s, 3H), 0.74 (s, 3H); 13C NMR (100 MHz,
CDCl3) 208.1, 196.8, 176.3, 133.3, 133.1, 133.0, 122.7, 120.2,
120.1, 107.9, 64.2, 63.9, 56.0, 45.0, 41.7, 41.0, 29.7, 28.7, 26.0,
1
oil; FTIR (CHCl3) 2866, 1665, 1461, 1384, 1366, 883 cm-1; H
NMR (500 MHz, CDCl3) 5.30 (br t, J ) 6.5, 1H), 5.10 (br t, J )
7.0, 1H), 4.69 (m, 1H), 2.38 (t, J ) 6.0, 2H), 2.18 (br d, J ) 14.0,
1H), 2.03 (dtd, J ) 17.0, 5.5, 2.0, 1H), 1.98 (m, 1H), 1.74 (m,
1H), 1.70 (s, 3H), 1.66 (s, 3H), 1.64 (m, 1H), 1.61 (s, 3H), 1.60 (s,
3H), 1.24 (m, 1H), 1.18 (m, 3H), 1.09 (m, 18H), 0.99 (s, 3H), 0.75
(s, 3H); 13C NMR (125 MHz, CDCl3) 152.3, 131.5, 128.3, 127.8,
124.5, 100.3, 51.1, 44.9, 36.8, 28.3, 27.1, 27.0, 26.0, 25.9, 25.7,
18.2, 18.1, 17.9, 17.8, 16.1, 13.1; HRMS (ESI) m/z calcd for C27H50-
OSi 419.3709, found 419.3690 [M + H]+.
(()-(1S,5R,7S)-2-Methoxy-8,8-dimethyl-1,7-bis(prenyl)bicyclo-
[3.3.1]non-2-ene-4,9-dione (22) and (()-(1S,5R,7S)-4-Methoxy-
8,8-dimethyl-1,7-bis(prenyl)bicyclo[3.3.1]non-3-ene-2,9-dione (23).
To a solution of enol ethers 16 and 17 (470 mg, 1.70 mmol) in
Et2O (1.5 mL) at -20 °C (dry ice/CCl4 bath) was added dropwise
malonyl dichloride (165 µL, 1.70 mmol, 1 equiv). The reaction
mixture was stirred at -20 °C for 24 h, and then benzyltriethy-
lammonium chloride (19 mg, 85 µmol, 0.05 equiv) was added,
followed by a solution of KOH (381 mg, 6.8 mmol, 4 equiv) in
H2O (1.5 mL). The reaction mixture was allowed to reach rt and
stirred for 5 h. After dilution with water (25 mL) and petroleum
ether (25 mL), the pH was adjusted to ∼10 using 2 N NaOH. The
phases were separated, and the aqueous phase was extracted using
petroleum ether (2 × 25 mL). The combined organic phases were
dried over MgSO4, and the solvent was evaporated. The residue
obtained was purified by column chromatography (eluent petroleum
ether/AcOEt 95/5) to provide ketones 14 and 15 as an 85/15 mixture
(260 mg, 58%). The aqueous layer was cooled to 0 °C, carefully
4812 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
25.9, 25.8, 24.2, 23.9, 18.0, 17.9, 17.8, 16.0; HRMS (ESI) m/z calcd
for C27H41O3 413.3061, found 413.3044 [M + H]+.
inseparable mixture of regioisomers (colorless oil, 4.13 g, 98%):
FTIR (CHCl3) 2927, 2856, 1713, 1639, 1471, 1369, 1189, 1006,
1
939, 819, 831 cm-1; H NMR (400 MHz, CDCl3) 5.81-5.70 (m,
(()-(1S,5R,7S)-3-Benzoyl-2-methoxy-8,8-dimethyl-1,5,7-tris-
(prenyl)bicyclo[3.3.1]non-2-ene-4,9-dione (29). A LTMP solution
(0.5 M, 824 µL, 412 µmol, 2 equiv), prepared from nBuLi
(312 µL, 1.6 M in hexanes, 0.5 mmol) and 2,2,6,6-tetramethylpi-
peridine (85 µL, 0.5 mmol) in THF (600 µL) at 0 °C, was added
to a -78 °C solution of 28 (85 mg, 206 µmol) in THF (4 mL).
The solution was stirred for 35 min, and then benzoyl chloride
(71.5 µL, 618 µmol, 3 equiv) was added and the resulting solution
stirred for 30 min at -78 °C. The mixture was quenched using
saturated aqueous NH4Cl and allowed to reach rt. The layers were
separated, and the aqueous phase was extracted using Et2O. The
combined organic phases were dried over MgSO4 and concentrated
to an oil, which was purified by column chromatography (eluent
petroleum ether/AcOEt 95/5 to 9/1) to give 29 (95 mg, 89%): Rf
) 0.44 (petroleum ether/AcOEt 9/1); white solid; mp ) 91-95
1H), 5.02-4.94 (m, 3H), 4.79-4.76 (m, 1H), 2.34-2.26 (m, 1H),
2.15-2.07 (m, 1H), 1.93-1.86 (m, 1H), 1.75-1.65 (m, 3H), 1.32-
1.25 (m, 1H), 0.96 (s, 3H), 0.91 (s, 9H), 0.85 (s, 3H), 0.16 (s, 6H);
13C NMR (125 MHz, CDCl3) 149.2, 138.6, 115.2, 102.5, 44.5, 41.9,
34.0, 33.0, 28.8, 26.7, 24.0, 23.8, 22.0, -2.9; HRMS (ESI) m/z
calcd for C17H33OSi 281.2301, found 281.2306 [M + H]+.
(()-(1S,5R,7S)-7-Allyl-4-methoxy-8,8-dimethylbicyclo[3.3.1]-
non-3-ene-2,9-dione (55) via 54. Malonyl dichloride (0.70 mL,
14.2 mmol, 2 equiv) was added to a solution of the enol ether 53
(2.00 g, 7.11 mmol) in Et2O (4 mL) at -20 °C (dry ice/CCl4 bath).
The reaction mixture was stirred at -20 °C for 24 h, and then a
solution of BnEt3NCl (81 mg, 0.35 mmol, 0.05 equiv) and KOH
(3.19 g, 57.9 mmol, 8 equiv) in H2O (6 mL) was added dropwise.
The reaction mixture was allowed to reach rt over 6 h. After dilution
with H2O (25 mL) and petroleum ether (25 mL), the pH was
adjusted to ∼12 using 1 M KOH. The phases were separated, and
the aqueous phase was extracted using petroleum ether (2 ×
25 mL). The combined organic phases were dried over MgSO4 and
then concentrated in vacuo. Purification by column chromatography
gave ketone 52 as a clear yellow oil (567 mg, 48%). The aqueous
layer was then acidified to pH ∼1 using 2 M HCl and extracted
using CH2Cl2 (3 × 20 mL). The combined organic layers were
dried over MgSO4 and concentrated in vacuo to give 54 as a light
yellow foam (705 mg), used without purification in the next step:
FTIR (CHCl3) 3205, 2873, 1738, 1613, 1586, 1538, 1137,
°C; FTIR (CHCl3) 2914, 1726, 1675, 1645, 1587, 1376, 1318 cm-1
;
1H NMR (400 MHz, CDCl3) 7.81 (d, J ) 7.5, 2H), 7.54 (t, J )
7.5, 1H), 7.42 (t, J ) 7.5, 2H), 5.04 (m, 2H), 4.92 (m, 1H), 3.59
(s, 3H), 2.70 (br d, J ) 13.5, 1H), 2.57 (dd, J ) 13.5, 8.8, 1H),
2.42 (d, J ) 7.2, 2H), 2.14 (br d, J ) 12.5, 1H), 1.99 (m, 2H),
1.72 (m, 1H), 1.67 (s, 9H), 1.63 (s, 3H), 1.55 (s, 6H), 1.38 (t, J )
14.0, 1H), 1.16 (s, 3H), 0.79 (s, 3H); 13C NMR (125 MHz, CDCl3)
207.6, 196.3, 195.5, 173.4, 137.9, 134.1, 133.5, 133.4, 133.3, 129.2,
128.7, 123.3, 122.5, 120.5, 119.9, 65.7, 64.1, 60.5, 46.1, 41.8, 41.4,
29.5, 28.7, 26.0, 25.9, 25.8, 25.1, 24.1, 18.1, 18.0, 17.9, 16.3; HRMS
(ESI) m/z calcd for C34H45O4 517.3323, found 517.3314 [M + H]+.
(()-Clusianone from Derivative 29. To a solution of 29
(18 mg, 9.3 µmol) in DMSO (1 mL) was added LiCl (3 mg, 70
µmol, 8 equiv), and the resulting solution was stirred at 120 °C for
2 h. The mixture was cooled at rt, diluted with H2O (3 mL), and
extracted using Et2O. The combined organic phases were dried over
MgSO4 and concentrated in vacuo. Column chromatography (eluent
petroleum ether/AcOEt 4/1) afforded (()-clusianone as white solid
(10 mg, 57%). Data were identical to those previously reported,
and a detailed comparison of our NMR data with those previously
reported can be found in the Supporting Information associated with
our earlier communication.15
1
852 cm-1; H NMR (400 MHz, CDCl3) 5.87 (s, 1H), 5.71-5.61
(m, 1H), 5.05-4.80 (m, 2H), 3.16-3.14 (m, 1H), 2.83-2.81 (m,
1H), 2.35-2.28 (m, 1H), 2.20-2.13 (m, 1H), 1.88-1.78 (m, 1H),
1.76-1.60 (m, 2H), 1.15 (s, 3H), 0.90 (s, 3H); 13C NMR (125 MHz,
CDCl3) 206.2, 190.2, 186.3, 136.4, 117.0, 107.7, 68.9, 57.3, 42.0,
38.1, 33.8, 32.4, 27.1, 20.5; HRMS (ESI) m/z calcd for C14H19O3
235.1329, found 235.1329 [M + H]+. The crude bicyclic product
54 (1.20 g) was dissolved in CH3OH (20 mL), p-TsOH (97 mg,
10 mol %) was added, and the solution then heated under reflux
for 10 h. The mixture was allowed to cool and concentrated in
vacuo. Purification by column chromatography (petroleum ether/
EtOAc 8/2) afforded 55 (996 mg, 29% from enol ether 53) as a
yellow solid: mp 129-131 °C; Rf ) 0.24 (petroleum ether/AcOEt
4/1); FTIR (CHCl3) 3678, 2928, 1736, 1650, 1603, 1373, 1348,
4-Allyl-3,3-dimethylcyclohexanone (52). Methylmagnesium
bromide (17.1 mL, 3 M in Et2O, 1.3 equiv) was slowly cannulated
into a solution of enone 51 (5.93 g, 0.039 mol) and copper iodide
(0.376 g, 1.95 mmol, 0.05 equiv) in THF/DMS (100/10 mL) at 0
°C under N2. The reaction mixture was stirred at 0 °C for 15 min
and then poured into a mixture of saturated aqueous NH4Cl/NH4-
OH (1/1, 50 mL) and Et2O (80 mL). The layers were separated,
and the aqueous layer was extracted with Et2O (3 × 50 mL). The
combined organic extracts were dried over MgSO4 and concentrated
in vacuo to give a dark yellow oil. Purification by column
chromatography (petroleum ether/EtOAc 97/3 to 95/5) gave the
title compound 52 as a pale yellow oil (4.88 g, 74%): Rf ) 0.81
(petroleum ether/EtOAc 4/1); FTIR (CHCl3) 3678, 3599, 2926,
1
1223, 1185, 1001, 919, 858 cm-1; H NMR (400 MHz, CDCl3)
5.69 (s, 1H), 5.70-5.61 (m, 1H), 5.03-4.99 (m, 2H), 3.77 (s, 3H),
3.20-3.16 (m, 1H), 2.79-2.78 (m, 1H), 2.34-2.30 (m, 1H), 0.88
(s, 3H), 1.12 (s, 3H), 1.78-1.62 (m, 3H); 13C NMR (125 MHz,
CDCl3) 205.9, 193.9, 175.8, 136.6, 116.8, 105.3, 73.9, 56.9, 53.1,
42.6, 38.5, 33.5, 31.9, 26.8, 20.5; HRMS (ESI) m/z calcd for
C15H21O3 249.1491, found 249.1485 [M + H]+.
(()-(1S,5R,7S)-7-Allyl-4-methoxy-8,8-dimethyl-5-prenylbicyclo-
[3.3.1]non-3-ene-2,9-dione (56). A solution of LDA‚LiCl was
prepared by treatment of a suspension of DIPA‚HCl (472 mg,
3.42 mmol, 5 equiv) in THF (5 mL) at -78 °C with nBuLi (1.6 M
solution in hexanes; 4.3 mL, 6.85 mmol). The solution was allowed
to warm to rt and after 10 min recooled to -78 °C. This LDA‚
LiCl solution was added dropwise via syringe to a solution of
bicycle 55 (170 mg, 0.68 mmol) in THF (1 mL) at -78 °C, resulting
in a deep yellow solution. The solution was stirred at -78 °C for
35 min and then quenched with prenyl bromide (0.79 mL, 10 equiv)
and stirred for a further 2.5 h at -78 °C. The reaction mixture was
quenched after this period with H2O (5 mL) followed by extraction
with EtOAc (3 × 5 mL). The organic layers were combined and
washed with saturated aqueous NaCl (5 mL), dried (MgSO4), and
concentrated in vacuo. Purification by column chromatography
(petroleum ether/AcOEt 4/1) gave the title compound 56 as a yellow
solid (115 mg, 0.36 mmol, 53%): mp 90-92 °C; Rf ) 0.53
(petroleum ether/AcOEt 4/1); FTIR (CHCl3) 2932, 2857, 1731,
1650, 1592, 1454, 1395, 1371, 1112, 1057, 983, 980, 916,
1
1713, 1605, 1075, 1006, 838 cm-1; H NMR (500 MHz, CDCl3)
5.85-5.76 (m, 1H), 5.08-5.03 (m, 2H), 2.46-2.40 (m, 1H), 2.37-
2.23 (m, 3H), 2.11 (dd, J ) 13.5, 2.2, 1H), 2.09-2.04 (m, 1H),
1.79-1.71 (m, 1H), 1.61 (tt, J ) 11.0, 3.1, 1H), 1.54-1.44 (m,
1H), 1.07 (s, 3H), 0.82 (s, 3H); 13C NMR (125 MHz, CDCl3) 212.0,
137.8, 116.2, 55.8, 44.9, 40.7, 38.5, 34.0, 29.8, 27.4, 20.8; HRMS
(ESI) m/z calcd for C11H19O 167.1436, found 167.1430 [M + H]+.
1-tert-Butyldimethylsilyloxy-4-allyl-5,5-dimethylcyclohex-1-
ene (53). Sodium iodide (4.3 g, 0.028 mol, 2 equiv), tert-
butyldimethylsilyl chloride (4.3 g, 0.028 mol, 2 equiv), and freshly
distilled triethylamine (8.0 mL, 0.057 mol, 4.0 equiv) were added
to a solution of cyclohexanone 52 (2.37 g, 0.014 mol) in CH3CN
(30 mL). The reaction mixture was heated under reflux for 4.5 h
and then cooled to rt and concentrated to give a brown solid. After
trituration of the solid with petroleum ether and concentration in
vacuo, a pale yellow oil was obtained which was purified by column
chromatography (1% Et3N in petroleum ether) to give 53 as a 5:1
1
838 cm-1; H NMR (500 MHz, CDCl3) 5.73 (s, 1H), 5.70-5.61
J. Org. Chem, Vol. 72, No. 13, 2007 4813

Ahmad et al.
(m, 1H), 5.04-4.94 (m, 3H), 3.75 (s, 3H), 2.85-2.84 (m, 1H),
2.47 (dd, J ) 14.4, 7.7, 1H), 2.37 (dd, J ) 14.4, 7.7, 1H), 2.32-
2.24 (m, 1H), 1.96 (dd, J ) 13.9, 4.4, 1H), 1.81-1.73 (m, 1H),
1.73-1.67 (m, 1H), 1.64 (s, 3H), 1.63 (s, 3H), 1.38 (t, J ) 12.5,
1H), 1.10 (s, 3H), 0.75 (s, 3H); 13C NMR (125 MHz, CDCl3) 206.0,
193.6, 177.7, 136.8, 133.8, 119.2, 116.6, 106.1, 74.6, 57.0, 56.8,
42.7, 39.7, 39.1, 33.6, 29.6, 26.5, 25.9, 20.6, 17.9; HRMS (ESI)
m/z calcd for C20H29O3 317.2117, found 317.2111 [M + H]+.
(()-(1S,3S,8S,10S)-10-Allyl-3-(1-hydroxy-1-methylethyl)-9,9-
dimethyl-4-oxatricyclo[6.3.1.01,5]dodec-5-ene-7,12-dione (57) and
(()-(1S,3R,8S,10S)-10-Allyl-3-(1-hydroxy-1-methylethyl)-9,9-
dimethyl-4-oxatricyclo[6.3.1.01,5]dodec-5-ene-7,12-dione (58). A
solution of dimethyldioxirane (9.5 mL, ∼0.05 M in acetone, 1.5
equiv) was added to substrate 56 (100 mg, 0.31 mmol) in CH2Cl2
(4 mL) at 0 °C, and the reaction mixture was stirred at 0 °C for
2.5 h. MgSO4 was then added to the solution which was filtered
and then concentrated in vacuo. The crude yellow solid was
redissolved in CH2Cl2 (4 mL) and cooled to 0 °C, and then
trimethylsilyl chloride (0.08 mL, 0.61 mmol, 2 equiv) was added
and the reaction further stirred for 2.5 h. The reaction was quenched
with NaHCO3 (8 mL), and the organic layer was separated and
washed with saturated NaCl (8 mL), extracted with CH2Cl2 (2 ×
10 mL), dried over MgSO4, and concentrated in vacuo. Purification
by column chromatography (petroleum ether/AcOEt 85/15) gave
first 58 as a clear, colorless oil (37 mg, 38%) and second 57 also
as a clear, colorless oil (37 mg, 38%). Data for 58: Rf ) 0.63
(petroleum ether/AcOEt 1/1); FTIR (CHCl3) 3600, 2922, 2856,
26.6, 26.6, 25.7, 20.5, 2.3; HRMS (IC) m/z calcd for C22H35O4Si
391.2305, found 391.2312 [M + H]+. Data for 60: Rf ) 0.30
(petroleum ether/EtOAc 95/5); FTIR (CHCl3) 2922, 1732, 1643,
1
1635, 1373, 1199, 1046, 1008, 844 cm-1; H NMR (500 MHz,
CDCl3) 5.79 (s, 1H), 5.75-5.64 (m, 1H), 5.07-4.97 (m, 2H), 4.18
(dd, J ) 10.2, 5.6, 1H), 2.94 (dd, J ) 13.2, 10.2, 1H), 2.82 (s,
1H), 2.31 (dd, J ) 13.8, 4.8, 1H), 2.36-2.29 (m, 1H), 1.91-1.71
(m, 3H), 1.48 (dd, J ) 13.7, 12.2, 1H), 1.40 (s, 3H), 1.23 (s, 3H),
1.17 (s, 3H), 0.88 (s, 3H), 0.15 (s, 9H); 13C NMR (125 MHz,
CDCl3) 204.0, 194.3, 179.3, 136.8, 116.8, 105.0, 93.2, 73.5, 72.8,
59.7, 41.9, 40.7, 40.3, 33.4, 27.2, 27.0, 26.7, 26.4, 20.4, 2.3; HRMS
(CI) m/z calcd for C22H35O4Si 391.2305, found 391.2299 [M +
H]+.
(()-(1S,3S,8S,10S)-6,10-Diallyl-9,9-dimethyl-3-(1-methyl-1-
trimethylsilanyloxyethyl)-4-oxatricyclo[6.3.1.01,5]dodec-5-ene-7,-
12-dione (61). A solution of LTMP was prepared by adding nBuLi
(0.208 mL, 1.6 M in hexanes, 0.33 mol, 2 equiv) to a solution of
tetramethylpiperidine (0.056 mL, 0.33 mmol, 2 equiv) in THF
(0.6 mL) at -78 °C. The solution was stirred at 0 °C for 15 min
then recooled to -78 °C. The LTMP solution was then added to
the tricyclic compound 59 (65 mg, 0.167 mmol) in THF (0.6 mL)
at -78 °C and stirred at that temperature for 20 min. After this
period, freshly prepared Li(2-Th)CuCN solution (1.3 mL, 0.25 M,
2 equiv) was added to the reaction mixture, which was then stirred
at -40 °C for 30 min. The reaction mixture was recooled to
-78 °C, and allyl bromide (0.07 mL, 0.83 mmol, 5 equiv) was
added dropwise. The solution was allowed to warm from -78 to
0 °C over 4 h. The reaction mixture was quenched by pouring it
into a mixture of saturated aqueous NH4Cl/NH4OH (1:1, 3 mL)
and Et2O (5 mL). The layers were separated, and the aqueous layer
was extracted with Et2O (3 × 5 mL). The combined organic extracts
were dried over MgSO4 and concentrated in vacuo to a viscous oil
which was purified by column chromatography (petroleum ether/
EtOAc 9/1) to give the title compound 61 as an opaque colorless
oil (36 mg, 0.084 mmol, 50%): Rf ) 0.55 (petroleum ether/EtOAc
95/5); FTIR (CHCl3) 2925, 1732, 1623, 1371, 1176, 1050,
844 cm-1; 1H NMR (500 MHz, CDCl3) 5.81 (s, 1H), 5.66 (s, 1H),
5.06-5.00 (m, 3H), 4.96-4.92 (m, 1H), 4.20 (dd, J ) 10.3, 5.8,
1H), 3.12-3.01 (m, 2H), 2.87 (s, 1H), 2.65 (dd, J ) 13.0, 10.3,
1H), 2.37-2.29 (m, 1H), 2.10 (dd, J ) 13.7, 4.5, 1H), 1.82-1.68
(m, 3H), 1.48 (dd, J ) 13.7, 12.1, 1H), 1.29 (s, 3H), 1.24 (s, 3H),
1.14 (s, 3H), 0.89 (s, 3H), 0.09 (s, 9H); 13C NMR (125 MHz,
CDCl3) 204.0, 193.0, 175.0, 136.7, 135.4, 116.9, 114.9, 114.1, 90.9,
73.9, 73.2, 59.7, 42.3, 39.4, 38.1, 33.5, 29.6, 27.2, 26.7, 26.4, 25.7,
20.5, 2.3; HRMS (ESI) m/z calcd for C25H39O4Si 431.2617, found
431.2624 [M + H]+.
1
1731, 1649, 1625, 1373, 1199, 1120, 1045, 1008, 831 cm-1; H
NMR (500 MHz, CDCl3) 5.80 (s, 1H), 5.72-5.63 (m, 1H), 5.04-
4.99 (m, 2H), 4.34 (dd, J ) 10.4, 4.8, 1H), 2.98 (dd, J ) 13.4,
10.4, 1H), 2.85-2.82 (s, 1H), 2.51 (dd, J ) 14.1, 4.6, 1H), 2.34-
2.29 (m, 1H), 1.89-1.67 (m, 3H), 1.46 (dd, J ) 14.1, 12.5, 1H),
1.42 (s, 3H), 1.24 (s, 3H), 1.16 (s, 3H), 0.89 (s, 3H); 13C NMR
(125 MHz, CDCl3) 203.8, 194.2, 179.0, 136.6, 116.8, 105.6, 91.9,
72.8, 70.9, 59.5, 41.9, 40.1, 40.0, 33.3, 27.4, 27.3, 26.6, 25.6, 20.3;
HRMS (ESI) m/z calcd for C19H27O4 319.1909, found 319.1904
[M + H]+. Data for 57: Rf ) 0.49 (petroleum ether/AcOEt 1/1);
FTIR (CHCl3) 3589, 2922, 1734, 1708, 1650, 1627, 1518, 1362,
1
1176, 920, 849 cm-1; H NMR (500 MHz, CDCl3) 5.77 (s, 1H),
5.72-5.63 (m, 1H), 5.70-5.00 (m, 2H), 4.58 (dd, J ) 11.0, 5.5),
2.83 (s, 1H), 2.63 (dd, J ) 13.0, 11.0), 2.39-2.33 (m, 1H), 2.13
(dd, J ) 13.7, 4.6, 1H), 1.91 (br s, 1H), 1.89-1.82 (m, 1H), 1.78-
1.68 (m, 2H), 1.52 (t, J ) 13.7, 1H), 1.35 (s, 3H), 1.20 (s, 3H),
1.15 (s, 3H), 0.9 (s, 3H); 13C NMR (125 MHz, CDCl3) 203.9, 194.3,
178.8, 136.5, 117.1, 104.4, 90.9, 73.1, 70.8, 59.8, 42.5, 39.2, 37.5,
33.4, 29.3, 26.6, 26.6, 24.1, 20.5; HRMS (ESI) m/z calcd for
C19H27O4 319.1909, found 319.1897 [M + H]+.
(()-(1S,3S,8S,10S)-6,10-Diallyl-3-(1-hydroxy-1-methylethyl)-
9,9-dimethyl-4-oxatricyclo[6.3.1.01,5]dodec-5-ene-7,12-dione (48).
Triethylamine trihydrofluoride (0.68 mL, 4.18 mmol, 20 equiv) was
added to diallyl 61 (9 mg, 0.21 mmol) in THF (1.5 mL) at rt and
the resulting solution stirred for 2 h. Concentration in vacuo
followed by column chromatography (petroleum ether/EtOAc 4/1
to 7/3) afforded the title compound as a clear, colorless oil (7 mg,
93%): Rf ) 0.84 (petroleum ether); FTIR (CHCl3) 2934, 1735,
(()-(1S,3S,8S,10S)-10-Allyl-9,9-dimethyl-3-(1-methyl-1-trim-
ethylsilanyloxyethyl)-4-oxatricyclo[6.3.1.01,5]dodec-5-ene-7,12-
dione (59) and (()-(1S,3R,8S,10S)-10-Allyl-9,9-dimethyl-3-(1-
methyl-1-trimethylsilanyloxyethyl)-4-oxatricyclo[6.3.1.01,5]dodec-
5-ene-7,12-dione (60). Imidazole (160 mg, 2.35 mmol, 5 equiv),
DMAP (6 mg, 10 mol %), and trimethylsilyl chloride (0.30 mL,
2.35 mmol, 5 equiv) were added to the tricyclic compounds 57
and 58 (150 mg, 1:2 diastereoisomeric mixture, 470 µmol) in DMF
(5 mL) at 0 °C. The reaction mixture was stirred for 2 h at 0 °C
after which time the reaction was complete by TLC. Purification
by column chromatography (10% EtOAc in petroleum ether)
afforded first the title compound 59 as a cream-colored oil
(113 mg, 62%) and then the title compound 60 as as a clear oil (52
mg, 28%). Data for 59: Rf ) 0.33 (petroleum ether/EtOAc 95/5);
FTIR (CHCl3) 2920, 1731, 1642, 1620, 1048, 859 cm-1; 1H NMR
(500 MHz, CDCl3) 5.75 (s, 1H), 5.73-5.63 (m, 1H), 5.07-5.01
(m, 2H), 4.47 (dd, J ) 10.3, 5.7, 1H), 2.83 (s, 1H), 2.66 (dd, J )
13.0, 10.4, 1H), 2.39-2.33 (m, 1H), 2.10 (dd, J 13.8, 4.6, 1H),
1.92-1.84 (m, 1H), 1.76-1.68 (m, 2H), 1.51 (dd, J ) 13.6, 12.5,
1H), 1.31 (s, 3H), 1.23 (s, 3H), 1.17 (s, 3H), 0.90 (s, 3H), 0.08 (s,
9H); 13C NMR (125 MHz, CDCl3) 204.1, 194.4, 179.4, 136.6,
116.9, 104.0, 91.5, 73.6, 73.2, 59.7, 42.3, 39.3, 38.0, 33.5, 29.1,
1
1629, 1365, 992, 887 cm-1; H NMR (500 MHz, CDCl3) 5.84-
5.63 (m, 2H), 5.09-4.95 (m, 4H), 4.57 (dd, J ) 10.8, 5.7, 1H),
3.15 (ddt, J ) 14.8, 5.7, 1.7, 1H), 3.05 (dd, J ) 14.8, 7.0, 1H),
2.90 (s, 1H), 2.67 (dd, J ) 12.7, 10.8, 1H), 2.38-2.33 (m, 1H),
2.03 (dd, J ) 13.7, 4.4, 1H), 1.79-1.68 (m, 3H), 1.48 (dd, J 13.7,
12.2, 1H), 1.35 (s, 3H), 1.18 (s, 3H), 1.14 (s, 3H), 0.90 (s, 3H);
13C NMR (125 MHz, CDCl3) 203.7, 192.9, 174.5, 136.6, 135.6,
117.0, 114.8, 114.3, 90.3, 73.2, 70.9, 59.9, 42.4, 39.5, 37.7, 33.4,
29.6, 27.1, 26.8, 26.7, 23.9, 20.5; HRMS (ESI) m/z calcd for
C22H31O4 359.2222, found 359.2216 [M + H]+.
(()-(4S,6S)-4-Allyl-3-methyl-6-prenylcyclohex-2-enone (62).
A solution of LDA was prepared by adding nBuLi (28.7 mL,
i
2.5 M in hexanes, 0.046 mol, 1.1 equiv) to a solution of Pr2NH
(6.42 mL, 0.046 mol, 1.1 equiv) in THF (150 mL) at -78 °C. The
solution was stirred at 0 °C for 15 min and then recooled to
4814 J. Org. Chem., Vol. 72, No. 13, 2007

Synthesis of Polyprenylated Acylphloroglucinols
-78 °C. A solution of enone 51 (6.22 g, 0.041 mol) in THF
(25 mL) was then slowly cannulated into the freshly prepared LDA
solution. The resulting solution was stirred at -78 °C for 15 min,
0 °C for 30 min, and then recooled to -78 °C. Prenyl bromide
(6.16 mL, 0.053 mol, 1.3 equiv) was added dropwise, and the
solution was allowed to warm to rt over 3 h. The reaction mixture
was quenched with saturated NH4Cl, the layers were separated, and
the aqueous phase was extracted with Et2O (3 × 100 mL). The
combined organic phases were dried over MgSO4 and then
concentrated in vacuo to a dark yellow oil. Purification by column
chromatography (petroleum/AcOEt 9/1) gave 62 as a mixture of
diastereoisomers (4:1, pale yellow oil, 6.4 g, 71%): Rf ) 0.59
(petroleum ether/EtOAc 4/1); FTIR (CHCl3) 2860, 1666, 1378,
(125 MHz, CDCl3) 201.0, 200.9, 163.9, 162.8, 136.3, 135.1, 133.3,
133.2, 128.1, 126.6, 121.9, 121.7, 117.5, 117.1, 46.6, 41.6, 39.8,
38.9, 36.8, 35.6, 33.4, 31.2, 28.0, 27.5, 25.8, 25.8, 22.9, 21.9, 18.1,
18.0; HRMS (ESI) m/z calcd for C15H23O 219.1749, found 219.1748
[M + H]+.
Acknowledgment. We acknowledge DEFRA and a Marie-
Curie Fellowship for support of V.R. and GSK, Harlow, U.K.,
and the University of Nottingham for support of N.M.A.
Supporting Information Available: Full experimental data for
compounds 31-47 and 62-65. X-ray crystallographic data (ORTEP
1
and CIF) for compounds and copies of H NMR and 13C NMR
1
1107, 1052, 996, 922, 855 cm-1; H NMR (400 MHz, CDCl3)
spectra for all new compounds except 65. This material is available
free of charge via the Internet at http://pubs.acs.org.
5.90-5.66 (m, 2H), 5.14-5.03 (m, 3H), 2.55-2.45 (m, 1H), 2.43-
2.39 (m, 2H), 2.25-2.16 (m, 2H), 2.10-1.96 (m, 2H), 1.95 (s,
3H), 1.79-1.71 (m, 1H), 1.69 (s, 3H), 1.60 (s, 3H); 13C NMR
JO070388H
J. Org. Chem, Vol. 72, No. 13, 2007 4815