Enantiospecific, biosynthetically inspired formal total synthesis of (+)-liphagal

Article abstract of DOI:10.1021/ol100756z

A biosynthetically inspired synthesis of (+)-liphagal has been achieved from (+)-sclareolide in 13 steps (9% overall yield). The key step is a biomimetic ring expansion of a highly stabilized benzylic carbocation, which generates the seven-membered ring and the benzofuran of the natural product in a single cascade reaction.

Full text of DOI:10.1021/ol100756z

ORGANIC
LETTERS
2010
Vol. 12, No. 10
2394-2397
Enantiospecific, Biosynthetically
Inspired Formal Total Synthesis of
(+)-Liphagal
Jonathan H. George,* Jack E. Baldwin, and Robert M. Adlington
Department of Chemistry, Chemistry Research Laboratory, UniVersity of Oxford,
Mansfield Road, Oxford OX1 3TA, U.K.
jonathan.george@chem.ox.ac.uk
Received April 1, 2010
ABSTRACT
A biosynthetically inspired synthesis of (+)-liphagal has been achieved from (+)-sclareolide in 13 steps (9% overall yield). The key step is a
biomimetic ring expansion of a highly stabilized benzylic carbocation, which generates the seven-membered ring and the benzofuran of the
natural product in a single cascade reaction.
Liphagal (1) is a tetracyclic meroterpenoid natural product
isolated by Andersen et al.1 in 2006 during a program
designed to discover new inhibitors of the phosphoinositide-
3-kinase (PI3K) signaling pathway.2 A library of marine
invertebrate extracts were screened in a fluorescent polariza-
tion assay using human PI3K R expressed in SF9 insect cells,
and the MeOH extract of the Dominican marine sponge Aka
Coralliphaga showed promising activity. Liphagal (1) was
subsequently isolated from this extract, and its unusual
structure, featuring a 6-7 ring system fused to a benzofuran,
was assigned by comprehensive NMR studies. Liphagal (1)
showed inhibitory activity against PI3K R with an IC50 of
100 nM in the primary fluorescent polarization enzyme assay,
and it was observed to be approximately 10-fold more potent
against PI3K R than PI3K γ. In addition, liphagal was found
to be cytotoxic to LoVo (human colon, IC50 0.58 µM), CaCo
(human colon, IC50 0.67 µM), and MDA-468 (human breast,
IC50 1.58 µM) cell lines in secondary in Vitro cell assays.
Two possible biosynthetic pathways to form the fused 6-7
ring system of liphagal (1) were proposed by Andersen.1 In
the first pathway, polyene cyclization of a farnesylated
trihydroxybenzaldehyde I could give the tertiary carbocation
II, which contains a fused 6-6 ring system (Scheme 1). This
carbocation could be converted to the known natural product
siphonodictyal B3 (4), also isolated from Aka Coralliphaga,
via hydride shift and deprotonation. Siphonodictyal B (4)
could then be transformed into liphagal via epoxidation, ring
expansion of the resultant epoxide III to give ketone V,
epimerization at C-8 to give VI, and finally dehydration to
form the fused benzofuran of 1. Furthermore, we propose
that the ring expansion of III f V may occur via the
o-quinone methide IV in a process analogous to an R-hy-
droxy aldehyde rearrangement or a pinacol rearrangement.
o-Quinone methides have been invoked in a range of
biosynthetic pathways and used in several biomimetic
(1) (a) Marion, F.; Williams, D. E.; Patrick, B. O.; Hollander, I.; Mallon,
R.; Kim, S. C.; Roll, D. M.; Feldberg, L.; Van Soest, R.; Andersen, R. J.
Org. Lett. 2006, 8, 321. (b) Andersen, R.; Hollander, I.; Roll, D. M.; Kim,
S. C.; Mallon, R. G.; Williams, D. E.; Marion, F. Meroterpenoid Inhibitors
of Phosphoinositide 3 Kinase (PI3K). Patent WO2006081659, 2006-08-
10.
(3) (a) Sullivan, B.; Djura, P.; McIntyre, D.; Faulkner, D. J. Tetrahedron
1981, 37, 979. (b) Sullivan, B. W.; Faulkner, D. J. J. Org. Chem. 1986, 51,
4568.
(2) For a review of known PI3K inhibitors see Sundstrom, T. J.;
Anderson, A. C.; Wright, D. L. Org. Biomol. Chem. 2009, 7, 840.
10.1021/ol100756z  2010 American Chemical Society
Published on Web 04/15/2010

Scheme 1
.
Proposed Biosynthesis of Liphagal via Ring
Scheme 2. Proposed Biosynthesis of Corallidictyal A and B via
Expansion of an o-Quinone Methide Intermediate
Spirocyclization of a p-Quinone Methide Intermediate
syntheses of liphagal by Andersen1 and Mehta,7 who both
synthesized the brominated benzofuran 5. Acid-mediated
cyclization of this compound gave 6 in 40% yield as a 2:5
mixture of C-8 epimers in favor of the undesired ꢀ-isomer
(Scheme 3).
Scheme 3. Andersen’s Synthesis of (()-Liphagal
syntheses,4 although their rearrangement via alkyl shifts has
not been reported. However, in the case of liphagal biosyn-
thesis the intermediacy of an o-quinone methide seems likely.
Several stable o-quinone methides tethered to bicyclic
sesquiterpenoids similar to IV have been reported.5
As part of our continuing interest in the biomimetic
synthesis of marine derived natural products we proposed
to investigate the ring-expansion biosynthetic pathway to
liphagal, which we believe to be the more plausible biogen-
esis because of the co-isolation of siphonodictyal B (2) and
corallidictyals A (3) and B (4) from Aka Coralliphaga with
liphagal. Furthermore, this strategy could form the 6-7 fused
ring system from a more readily accessible 6-6 ring system
and could therefore lead to a more efficient synthesis of the
natural product that avoids the poor yield and low stereo-
selectivity of the direct polyene cyclization reported by
Andersen and Mehta.
The biosynthesis of liphagal via quinone methide inter-
mediates is supported by the co-isolation of corallidictyals
A (3) and B (4) from Aka Coralliphaga,6 which are
presumably formed by spirocyclization of the p-quinone
methide VII (Scheme 2).
The second biosynthetic pathway to liphagal proposed by
Andersen involves polyene cyclization of a benzofuran to
generate the 6-7 ring system directly.1 This more direct
pathway has been investigated in the course of biomimetic
(4) For a review of o-quinone methides in organic synthesis, see: Van
De Water, R. W.; Pettus, T. R. R. Tetrahedron 2002, 58, 5367.
(5) Barrero, A. F.; Alvarez-Manzaneda, E. J.; Chahboun, R.; Cortes,
M.; Armstrong, V. Tetrahedron 1999, 55, 15181.
Our proposed retrosynthesis of liphagal (1) is outlined in
Scheme 4. It was our belief that 1 could be generated from
(6) (a) Chan, J. A.; Freyer, A. J.; Carte, B. K.; Hemling, M. E.; Hofmann,
G. A.; Mattern, M. R.; Mentzer, M. A.; Westley, J. W. J. Nat. Prod. 1994,
57, 1543. (b) Grube, A.; Assmann, M.; Lichte, E.; Sasse, F.; Pawlik, J. R.;
Kock, M. J. Nat. Prod. 2007, 70, 504.
(7) Mehta, G.; Likhite, N. S.; Kumar, C. S. A. Tetrahedron Lett. 2009,
50, 5260.
Org. Lett., Vol. 12, No. 10, 2010
2395

Scheme 4. Biosynthetically Inspired Retrosynthesis of Liphagal
Scheme 6. Attempted Addition of an Aryllithium to R-Hydroxy
Aldehyde 19
the dimethoxy compound 7 via a formylation reaction
followed by demethylation. Benzofuran 7 could be formed
from ring expansion of the stabilized benzylic carbocation
9 (a synthetic equivalent of the o-quinone methide IV) via
a pinacol-type rearrangement to give ketone 8 followed by
dehydration. The stabilized carbocation intermediate 9 could
be generated under mild conditions from an o-hydroxy-
benzyl alcohol that could be formed by the acid-mediated
removal of the THP group from 10. It was envisaged that
the key ring-expansion substrate 10 could be synthesized
from the cheap, chiral starting material (+)-sclareolide (11).
Use of this enantiopure building block would ultimately
allow the absolute configuration of liphagal to be determined.
Our synthesis commenced with the conversion of (+)-
sclareolide (11) into diol 12 in three steps according to a
modified version of the procedure of Kuchkova et al. (Scheme
5).8 Swern oxidation of 12 gave the ꢀ-hydroxy aldehyde 13,9
which was dehydrated with BF3.Et2O in CH2Cl2 to give R,ꢀ-
unsaturated aldehyde 14. Reduction of this compound with
NaBH4 gave allylic alcohol 15, which was epoxidized with
mCPBA to give epoxides 16 and 17 as a 3:1 mixture (separable
by SiO2 flash column chromatography).10
Reductive ring opening of epoxide 16 with LiAlH4 gave
diol 18, which was subjected to a Swern oxidation to give
R-hydroxy aldehyde 19. The aryllithium generated from 2011
and t-BuLi in THF at -78 °C failed to add to 19 to give 10.
Instead, the main isolated product was the R-hydroxy ketone
22, which was formed in 40% yield (1.5:1 mixture of C-10
epimers). The ring expansion product 22 was presumably
formed by deprotonation of the hydroxy group of 19 to give
tertiary alkoxide 21, followed by an R-hydroxy aldehyde
rearrangement12 to give 22.
Although not the desired outcome, this reaction did at least
indicate selective migration of the more electron-rich
C(9)-C(10) bond instead of the C(8)-C(9) bond, as required
in our planned biomimetic synthesis of liphagal.
One approach to improve the addition of an aryllithium
species to the R-hydroxy-aldehyde 19 would be to protect
the hydroxy group. Alternatively, the order of the epoxide
ring opening, oxidation, and aryllithium addition could be
altered, as shown in Scheme 7. Oxidation of 16 with
Scheme 5. Synthesis of Epoxide 16 from (+)-Sclareolide
Scheme 7. Synthesis of Ring Expansion Precursor 10
2396
Org. Lett., Vol. 12, No. 10, 2010

Dess-Martin periodinane gave aldehyde 23, which reacted
smoothly with the aryllithium generated from 20 and t-BuLi
to give 24 in 94% yield. Reductive ring opening of the
epoxide of 24 with LiAlH4 then gave 10, which was used
without further purification.
With 10 in hand, the key biomimetic ring-expansion
reaction could be investigated. Treatment of crude 10 with
TFA in CH2Cl2 at -78 °C followed by gradual warming to
room temperature gave the ring-expanded product 7 in 74%
yield over two steps (Scheme 8).
10 f 8 proceeds via an o-quinone methide rather than a
stabilized carbocation.
The synthesis of (+)-liphagal was continued by ortho-
lithiation of 7 and quenching of the resultant aryllithium
species with DMF to install the aromatic aldehyde group of
26 (Scheme 9). This compound has previously been dem-
Scheme 9. Completion of the Synthesis of Liphagal
Scheme 8
.
Ring Expansion via Pinacol-Type Rearrangement of
a Stabilized Benzylic Carbocation
ethylated with BI3 by Andersen et al. to give 1. The 1H and
13C NMR spectra of 26 were identical to those previously
reported,1 as was the [R]D,1b which confirms that the absolute
configuration of (+)-liphagal is (5S,8R,11S).
In summary, a concise and enantiospecific formal total
synthesis of a potent PI3K inhibitor, liphagal (1), has been
achieved in 13 steps (9% overall yield) from (+)-sclareolide
(11). The key step is a biosynthetically inspired cascade
reaction involving ring expansion of a benzylic alcohol,
followed by benzofuran formation. Importantly, the late-stage
introduction of the aryl group in the synthesis should allow
access to a range of liphagal analogues with differentially
substituted benzofuran rings for evaluation as PI3K inhibi-
tors.
This cascade reaction presumably proceeds via initial
removal of the labile phenolic THP group to give 25, which
could undergo a facile dehydration to generate the highly
stabilized benzylic carbocation 9. Pinacol rearrangement of
this transient intermediate via selective migration of the
C(9)-C(10) bond would then give the cycloheptanone 8,
which could undergo dehydration to form the benzofuran of
7. Note that similar acidic reaction conditions have been used
to generate o-quinone methides from suitably substituted
benzylic alcohols,13 so it is possible that the ring expansion
Acknowledgment. We thank F. Hoffman-La Roche Ltd,
Basel, Switzerland for postdoctoral funding (J.G.).
Supporting Information Available: Experimental pro-
cedures and spectral data for compounds 7, 10, 15-19, 23,
24, and 26. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL100756Z
(11) Handy, S. T.; Zhang, Y.; Bregman, H. J. Org. Chem. 2004, 69,
2362.
(8) (a) Kuchkova, K. I.; Chumakov, Y. M.; Simonov, Y. A.; Bocelli,
G.; Panasenko, A. A.; Vlad, P. F. Synthesis 1997, 1045. (b) Vadapalli, S.;
Kane, C. T. Org. Prep. Proced. Int. 2008, 40, 201.
(9) Fujiwara, N.; Kinoshita, M.; Akita, H. Tetrahedron: Asymmetry 2004,
17, 3037.
(12) For a review see: Paquette, L. A.; Hofferberth, J. E. The a-Hydroxy
Ketone (a-Ketol) and Related Rearrangements. In Organic Reactions;
Overman, L. E., Ed.; John Wiley & Sons: New York, 2003; Vol. 63, pp
477-567.
(10) Kulcitki, V.; Ungur, N.; Gavagnin, M.; Carbone, M.; Cimino, G.
Eur. J. Org. Chem. 2005, 1816.
(13) Arjona, O.; Garranzo, M.; Mahugo, J.; Maroto, E.; Plumet, J.; Saez,
B. Tetrahedron Lett. 1997, 38, 7249.
Org. Lett., Vol. 12, No. 10, 2010
2397