A short biomimetic approach to the fully functionalized bicyclic framework of type A acylphloroglucinols

Article abstract of DOI:10.1021/ol901781n

A biomimetic approach toward type A polyprenylated acylphlorogluclnols (PPAPs) Is described. The method Is based on a C-alkylation-cation cyclization reaction sequence, leading to a convenient buildup of molecular complexity, employing the simple and read

Full text of DOI:10.1021/ol901781n

ORGANIC
LETTERS
2009
Vol. 11, No. 19
4430-4433
A Short Biomimetic Approach to the
Fully Functionalized Bicyclic Framework
of Type A Acylphloroglucinols
Elias A. Couladouros,* Marianna Dakanali, Konstantinos D. Demadis, and
Veroniki P. Vidali
Chemical Laboratories, Agricultural UniVersity of Athens, Iera Odos 75, GR-11855
Athens, Greece, Crystal Engineering, Growth & Design Laboratory, Department of
Chemistry, UniVersity of Crete, P.O. Box 2208, Heraklion, Crete GR-71003 Greece,
and Laboratory of Natural Products Synthesis and Bioorganic Chemistry, NCSR
“Demokritos”, Ag. ParaskeVi, Athens GR-15310, Greece
ecoula@aua.gr
Received August 3, 2009
ABSTRACT
A biomimetic approach toward type A polyprenylated acylphloroglucinols (PPAPs) is described. The method is based on a C-alkylation-cation
cyclization reaction sequence, leading to a convenient buildup of molecular complexity, employing the simple and readily available
deoxycohumulone and an appropriately functionalized hydroxy halide. Thus, a versatile construction of the fully functionalized bicyclic framework
of type A PPAPs (5) was achieved.
Polycyclic polyprenylated acylphloroglucinols (PPAPs), such
as hyperforin (1), nemorosone (2), garsubelin A (3), clu-
sianone (4), etc. (Scheme 1), have been attracting significant
attention of many research groups for over 20 years. Due to
their intriguing biological activities combined with their
unique molecular architecture, these molecules and their
analogues constitute extremely challenging synthetic targets
as well as very promising substrates for biological evalua-
tion.1 Grossman first classified all known PPAPs into three
main types (A-C) based on proposed biosynthetic path-
ways.2
Although many approaches for the synthesis of the
1-acylbicyclo[3.3.1]nonane-2,4,9-trione structure of type A
PPAPs have been reported, only few led successfully to a
total synthesis (garsubelin A and nemorosone).3,4 Moreover,
in most methods, installation of all oxygen functionalities
or side chains results in a significant increase of the required
synthetic steps.5,6 It is evident that the most challenging
features in the synthesis of PPAPs is the construction of an
extremely congested carbon framework containing three
quaternary carbon centers, two of which are on a bridgehead
position of the bicyclic system (C-1 and C-5), while the third
(C-8) is neighboring C-1.
Herein, we wish to demonstrate a general method leading
efficiently and in few steps to the fully functionalized acyl
bicyclo[3.3.1]nonane-2,4,9-trione core of type A PPAPs.
Inspired by the biosynthetic pathways proposed for type A
PPAPs,1 we aimed at the construction of key intermediate 5
(Scheme 2), employing an annulation reaction, which would
involve an intermolecular cation cyclization of colupulone
analogue 6, the latter being formed by C-allylation of
deoxycohumulone 7 with hydroxyallyl halide 8.
For the preparation of colupulone analogues 6, deoxyco-
humulone 7 was prepared by prenylation of acylphloroglu-
cinol 9 (Scheme 3) with dimethylallyl bromide 10.7 At a
first attempt, C-allylation of 7 with bromide 8a was
10.1021/ol901781n CCC: $40.75
Published on Web 09/09/2009
 2009 American Chemical Society

Scheme 1
.
Natural PPAPs
Scheme 3. C-Allylation of Deoxycohumulone 7
Accordingly, colupulone analogues 6b-d (Table 1) were
similarly prepared utilizing method A. It was observed that
when chlorides were used, improved yields of C-allylated
products 6 were obtained, probably due to their higher
stability.
Table 1. Allylation Studies: Preparation of Colupulone
Analogues 6b-d
Scheme 2. Retrosynthetic Analysis
examined. Although 6a was easily derived under alkaline
conditions in protic or aprotic solvents (methods B and C),
a significantly improved yield was achieved when a two-
phase solvent system at pH 14 was applied (method A).
a Product was formed as ∼1:1 mixture of diastereoisomers (1H NMR).
(1) For a recent review refering to biological activity, biosynthesis, and
synthetic efforts towards PPAPs, see: (a) Ciochina, R.; Grossman, R., B.
Chem. ReV. 2006, 106, 3963. For more recent studies towards the synthesis
of phloroglucinols, see: (b) Abe, M.; Nakada, M. Tetrahedron Lett. 2007,
48, 4873. (c) Kraus, A. G.; Jeon, I. Tetrahedron Lett. 2008, 49, 286. (d)
Mehta, G.; Bera, M. K. Tetrahedron Lett. 2009, 50, 3519. For the synthesis
of Hyperforin analogues and their biological evaluation, see: (e) Verotta,
L.; Appendino, G.; Jakupovic, J.; Bombardelli, E. J. Nat. Prod. 2000, 63,
412. (f) Verotta, L.; Appendino, G.; Belloro, E.; Bianchi, F.; Sterner, O.;
Lovati, M.; Bombardelli, E. J. Nat. Prod. 2002, 65, 433. (g) Verotta, L.;
Appendino, G.; Bombardelli, E.; Brun, R. Bioorg. Med. Chem. Lett. 2007,
17, 1544. For the synthesis of designed derivatives and their biological
properties, see: (h) Gurtner, M.; Simon, J. C.; Giannis, A.; Sleeman, J. P.
ChemBioChem 2005, 6, 171. (i) Grey, C.; Kyrylenco, S.; Hennig, L.;
Nguyen, L.-H. D.; Bu¨ttner, A.; Pham, H. D.; Giannis, A. Angew. Chem.,
Int. Ed. 2007, 46, 5219. (j) Rothley, M.; Schmid, A.; Thiele, W.; Schacht,
V.; Plaumann, D.; Gartner, M.; Yektaoglu, A.; Bruye`re, F.; Noe¨l, A.;
Giannis, A.; Sleeman, J. P. Int. J. Cancer 2009, 125, 34. For recent reviews
on the biological activity of hyperforin, see: (k) Quiney, C.; Billard, C.;
Salanoubat, C.; Fourneron, J. D.; Kolb, J. P. Leukemia 2006, 20, 1519. (l)
Medina, M. A.; Martinez-Poveda, B.; Amores-Sanchez, M. I.; Quesada,
A. R. Life Sci. 2006, 79, 105.
Proceeding to the annulation step, C-4 enol of 6a was
selectively acetylated prior to cyclization (Scheme 4).8
Direct generation of an intermediate cation on 11 by the
use of acids, such as PPTS, CSA, or TsOH, in various
solvents (CH2Cl2, toluene, NMP, THF) or Pd(PPh3)4 in THF
proved completely unsuccesful leading to no reaction or
degradation products. The same results were obtained when
indirect methods involving transformation of the tertiary
(4) For synthesis of a type B PPAP, clusianone, see: (a) Rodeschini,
V.; Ahmad, N. M.; Simpkins, N. S. Org. Lett. 2006, 8, 5283. (b) Nuhant,
P.; David, M.; Pouplin, T.; Marazano, C. Org. Lett. 2007, 9, 287. (c) Qi,
J.; Porco, J. A., Jr. J. Am. Chem. Soc. 2007, 129, 12682. (d) See also ref
3d.
(5) To the best of our knowledge, methods establishing all quaternary
centers and oxygen functionalities, during cyclization, have been developed
for type B clusianone only. For two examples, see ref 4b,c.
(6) For an attempt to construct the bicyclic framework of type A PPAPs,
in a direct way, under oxidative conditions which, nevertheless, led to
polycyclic systems, indicating the difficulty of this approach, see: Mitasev,
B.; Porco, J. A., Jr. Org. Lett. 2009, 11, 2285.
(2) For a first attempt on classification of polyprenylated benzoylphlo-
roglucinols, see: Cuesta-Rubio, O.; Velez-Castro, H.; Frontana-Uribe, B. A.;
Cadena, J. Phytochemistry 2001, 57, 279.
(3) For total syntheses of garsubelin A, see: (a) Kuramochi, A.; Usada,
H.; Yamatsugu, K.; Motomu, K.; Shibasaki, M. J. Am. Chem. Soc. 2005,
127, 14200. (b) Siegel, D. R.; Danishefsky, S., J. J. Am. Chem. Soc. 2006,
128, 1048. (c) Ahmad, N. M.; Rodeschini, V.; Simpkins, N. S.; Ward, S. E.;
Blake, A. J. J. Org. Chem. 2007, 72, 4803. For a total synthesis of
nemorosone and clusianone, see: (d) Tsukano, C.; Siegel, D. R.; Danishef-
sky, S. J. Angew. Chem., Int. Ed. 2007, 46, 840.
(7) US Patent 4,101,585, 1978.
(8) Cyclization attempts of unprotected colupulone 6a resulted either
in messy reactions or in O-cyclization of prenyl side chains due to
participation of the C-4 enol.
Org. Lett., Vol. 11, No. 19, 2009
4431

Scheme 5
Scheme 4. Annulation Studies Based on Cation Cyclization
hydroxyl group into a good leaving group (bromide, tosylate,
or triflate) and subsequent cyclization were attempted.
Gratifyingly, our efforts to convert the tertiary hydroxyl
group of 11 into the corresponding methanesulfonic ester
resulted in an unexpected rapid and direct transformation to
the cyclized product 5a. In addition to the desired C-cyclized
product 5a (path a), the formation of the O-cyclized product
12a was also observed (ratio 5a/12a 1.5:1, 89%), probably
due to attack of enol hydroxyl on C-6 to the intermediate
electrophilic carbon C-13 (path b). Moreover, since the same
cyclization products were obtained when methanesulfonic
anhydride was used, it is evident that the reaction pathway
proceeds via the formation of a nondetectable methane-
sulfonic ester,9 without the intermediacy of the corresponding
chloride.
method toward type A PPAPs. Consequently, deoxycohu-
mulone 7 was subjected to C-allylation with bromide 8h or
chloride 8i to furnish colupulone analogues 17 and 18,
respectively (Scheme 7).
Table 2. Annulation Studies: Preparation of Key Intermediates 5
To confirm structure of 5a, by single-crystal X-ray
diffraction, compound 13 was prepared from 6a using
p-bromobenzoyl as a protective group (Scheme 5). Crystal-
lization of the latter was unsuccesful; thus, it was reduced
to the more crystallizable derivative 14. The structure of 14
was unambiguously confirmed by X-ray analysis.10 In the
structure of 14 in Scheme 5 chiral carbons are highlighted
in black.
All other colupulone analogues 6b-d were similarly
acetylated, and their annulation was examined. The results
are summarized in Table 2. As expected, cation cyclization
of compounds 6b-d (entries 2-5) proceeded under the same
conditions leading to target intermediates 5b-d.
a C/O ratio was ∼1.5:1. b Overall yields of cyclized products. c Products
were isolated as 1:1 mixture of diastereoisomers (1H NMR).
We should note that the presence of a double bond, allylic
to the tertiary alcohol, is necessary for the stabilization of
the intermediate cation. Hence, any attempt to cyclize the
“saturated” colupulone analogue 15 to 16 (Scheme 6),11
either directly or indirectly (via the formation of an inter-
mediate tosylate), resulted only in elimination or other
degradation products.
Scheme 6
In parallel to the development of the above strategy,
Michael reaction was also examined as a possible annulation
After selective acetylation or pivaloylation, followed by
desilylation and subsequent oxidation of 19, 20, and 21, the
corresponding Michael acceptors 22, 23, and 24 were
prepared.
(9) Methanesulfonic ester could not be detected by TLC.
(10) See the Supporting Information for complete experimental details.
(11) In this case, for the preparation of 15, alkylation of 7 with bromide
8g was performed in MeONa/MeOH (method B), as method A resulted in
no reaction.
4432
Org. Lett., Vol. 11, No. 19, 2009

Scheme 7
.
Preparation of Michael Precursors
Table 3. Michael Approach Results to Type a Bicyclic
Framework
Several conditions were applied to induce an intermolecu-
lar Michael reaction (Table 3). It is worth noting that, during
our experimentation, Qi and Porco published a successful
application of a similar approach for the synthesis of type
B, clusianone.4c Nevertheless, in our case, only disubstituted
R,ꢀ-unsaturated aldehyde 22 cyclized to the desirable type
A bicyclic framework 25 (entry 1). In contrast, the more
substituted aldehyde 23 remained unreacted either under mild
or strong alkaline conditions in low temperatures (entry 2).
Further experimentation for cyclization of 23 or its more
stable pivaloate analogue 24 under harsh conditions led either
to adamantane derivative 27 (entry 4), in accordance with
Qi and Porco’s results,12 or aromatization products 26 or
28, due to removal of the allylic side chain (entries 3 and
5).
The above results confirm the difficulty in the construction
of the fully functionalized bicyclic core met in type A PPAPs.
Only colupulone analogues containing an allylic tertiary
alcohol led to successful cyclization, and although minor
variations, such as change of solvent, temperature, or salt
addition, did not improve the C/O product ratio, this
unprecenteded annulation step remains extremely advanta-
geous for the course of this project. Employing this method,
two quaternary centers were directly connected surpassing
problems faced by other approaches. Thus, the highly
stereochemically demanding systems of type A PPAPs,
hyperforin and garsubelin A, established with all oxygen
functionalities and quaternary carbon centers, were con-
a Conditions A: (a): NaHCO3, DMSO/H2O or K2CO3, DMSO or Li2CO3,
THF or Cs2CO3,THF or SiO2/CH2Cl2, 25-60 °C; (b): NaH, THF or
KHMDS, toluene, -78 to +15 °C. b Conditions B: piperidine, C6H6, 25
°C or NaHCO3, DMSO/H2O, 80 °C or KHMDS, toluene, 60 °C or LiHMDS,
THF, 25 °C.
structed in two key steps and an affordable overall yield.
Moreover, the presence of an allylic hydroxyl group in key
intermediates 5 provides suitable handle for further deriva-
tization, as well as for the preparation of a large variety of
acylphloroglucinol analogues useful for SAR studies. Current
efforts are devoted to further optimizing the cyclization
reaction, as well as expanding the scope of key intermediate
5.
Acknowledgment. We thank Professor Dr. A. Giannis
(University of Leipzig, Germany) for kindly providing high-
resolution mass spectra and pro-ACTINA S.A. for low-
resolution mass spectra.
Supporting Information Available: Experimental details
and characterization data for selected compounds, as well
as CIFs and single-crystal X-ray data for 14. This material
is available free of charge via the Internet at http://pubs.acs.org.
(12) Formation of 27 is due to deacetylation of 23, in situ Michael
addition on C-3 and subsequent aldol condensation, under the harsh
conditons applied. See ref 4c.
OL901781N
Org. Lett., Vol. 11, No. 19, 2009
4433