Synthetic studies toward A-74528

Article abstract of DOI:10.1021/ol103117j

A potentially biomimetic approach toward the complex polyketide A-74528 is described. It is based on highly substituted biaryl compounds, synthesized using advanced cross-coupling and condensation methodologies.

Full text of DOI:10.1021/ol103117j

ORGANIC
LETTERS
2011
Vol. 13, No. 6
1386–1389
Synthetic Studies toward A-74528
Anastasia Hager, Dmitry Mazunin, Peter Mayer, and Dirk Trauner*
€
Department of Chemistry and Pharmacology, Ludwig-Maximilians-Universitat,
Munich, Germany, and Center for Integrated Protein Science, 81377 Munich, Germany
dirk.trauner@lmu.de
Received January 11, 2011
ABSTRACT
A potentially biomimetic approach toward the complex polyketide A-74528 is described. It is based on highly substituted biaryl compounds,
synthesized using advanced cross-coupling and condensation methodologies.
A-74528 (1) is an unusual natural product that was
recently isolated from Streptomyces sp. SANK 61196 by
Ogita and co-workers (Figure 1). The compound was
found to activate the interferon system via inhibition of
20,50-oligoadenylate phosphodiesterase (20-PDE). Because
an overabundance of 20-PDE in tumor and viral cells
prevents their apoptosis, this provides a potential therapy
against cancer and viral infections.1
With its 30 carbons, A-74528 (1) is one of the most
complex and largest aromatic polyketides known to date.
Interestingly, the biosynthetic machinery responsible for
its formation is very similar to the type II polyketide
synthase that accounts for the formation of fredericamycin
(2) in Streptomyces griseus.2 This is surprising given the
considerable structural diversity between the two natural
products despite their identical carbon count.
Structurally, A-74528 consists of a hexacyclic core with
an appended R-pyrone moiety. Its unprecedented carbon
skeleton contains two acyl resorcinol motifs typical of type
II polyketides, which are flanking a perhydropyrene core.
This tetracyclic core features six contiguous stereocenters,
one of which is quaternary. The secondary alcohol on ring
A is in an unusual position for a polyketide, a fact that
Figure 1. Two biosynthetically related C30-polyketides from
Streptomyces sp.
needs to be accounted for by any proposed biosynthetic
pathway. Biosynthetically, A-74528 can be traced back
to a C30 epoxy polyketide 7, as shown in Scheme 1. This
epoxide presumably undergoes a series of condensa-
tions, additions, and substitutions to assemble the mole-
cular framework of the natural product.2 The exact
sequence of these ring formations, however, remains
unknown. It is conceivable, for instance, that A-74528
(1) directly stems from a strained enone 3, which
engages in a transannular Michael addition and epox-
ide opening, as indicated. Compound 3, in turn, could
arise from a dearomatizing intramolecular Michael
(1) Fujita, Y.; Kasuya, A.; Matsushita, Y.; Suga, M.; Kizuka, M.;
Iijima, Y.; Ogita, T. Bioorg. Med. Chem. Lett. 2005, 15, 4317–4321.
(2) Zaleta-Rivera, K.; Charkoudian, L. K.; Ridley, C. P.; Khosla., C.
J. Am. Chem. Soc. 2010, 132, 9122–9128.
r
10.1021/ol103117j
Published on Web 02/16/2011
2011 American Chemical Society

Scheme 1. A Biosynthetic and Retrosynthetic Analysis of
A-74528
Scheme 2. Synthesis of a Key Biaryl 14 via Suzuki Coupling
stereoselectivity. We found it fascinating to test the limits
of biomimetic synthesis and explore whether such a complex
pathway could be emulated in the laboratory. We initially
focused on aryl naphthalenes of type 4, which seem to be
more tractable than other polyketide intermediates postu-
lated in Scheme 1. This required advanced biaryl chemistry,
which is the subject of the present communication. We now
report the synthesis of a complex aryl naphthalene that has
most of the relevant features of the postulated biosynthetic
intermediate 4 and discuss our first attempts toward its
further cyclizations. These investigations initially focused
on challenging transition-metal catalyzed cross-couplings
and eventually led to a synthetic strategy based on poten-
tially biomimetic condensation chemistry.
Our first approach started with the known naphthalene
derivative 9, which is accessible in a lengthy yet practical
sequence from 3,5-dimethoxy benzaldehyde (8) (Scheme 2).4
Conversion of the benzylic alcohol 9 into the correspond-
ing bromide 10 followed by nucleophilic substitution with
cyanide gave the naphthalene derivative 11 in good yield.5 Its
addition involving aryl naphthalene 4.3 Alternatively,
A-74528 (1) could stem from a disrotatory 6π-electro-
cyclization of intermediate 5, followed by tautomeriza-
tion, conjugate addition, and attack on the epoxide.
Both 4 and 5 could be formed by intramolecular con-
densation from the partially aromatized polyketide 6,
which we consider to be a key intermediate in the
pathway. The question is whether this intermediate
first forms a 6-, 10-, 12-, or 14-membered ring. Com-
pound 6 itself presumably stems from epoxy polyketide
7, which in turn is assembled by the polyketide synthase
from acetyl and malonyl coenzyme A.
In Nature, it appears that these cyclizations are mediated
by various enzymatic activities in the type II polyketide
synthase, which determines their sequence and
(4) Piettre, A.; Chevenier, E.; Massardier, C.; Gimbert, Y.; Greene,
A. E. Org. Lett. 2002, 4, 3139–3142.
(3) For tandem Michael-Michael addition, see: Scheerer, J. R.;
Lawrence, J. F.; Wang, G. C.; Evans, D. A. J. Am. Chem. Soc. 2007,
129, 8968–8969.
(5) X-ray structures in cif format for compounds 11, 13, and 20 are
available at the Cambridge Chrystallographic Data Centre
(Registration numbers CCDC 795939, 795938, 795937).
Org. Lett., Vol. 13, No. 6, 2011
1387

Scheme 3. Biaryl Synthesis via Polyketide Condensation
Figure 2. X-ray structure of biaryl acid 20.
coupling partner, boronic ester 13,5 was obtained from
trihydroxybenzoic acid 12 in four steps following a known
protocol.6 Linkage of these two components through Suzuki
coupling required careful optimization. Eventually, we
settled on a protocol that involved microwave irradiation
and tetrakis(triphenylphosphine) palladium(0) as a catalyst.
Under these conditions, 11 could be coupled with 13 to
afford 14 as a racemic mixture of atropisomers (Scheme 2).
This is one of the few examples where a 3-fold o,o,o0-
substituted biaryl bond has been formed in acceptable yield
through transition-metal mediated cross-coupling.7
Although we were able to find a satisfactory solution for
the cross-coupling, we nevertheless felt that the pathway
taken was too long to support a total synthesis of A-74528
(1). Therefore, we explored an alternative pathway that
involves condensation chemistry and is also more biomi-
metic. It starts with the deprotonation of orsellinic acid
derivative 15, easily available on a large scale through
trimerization of methyl acetoacetate.8 Condensation of the
corresponding organolithium compound with Weinreb
amide 169 gave ketone 17. Upon treatment with DBU,
this material underwent enolization and intramolecular
condensation to afford isocoumarin 18, which has the
characteristics of an active ester. Indeed, 18 engaged in
another Claisen-type condensation with the anion of 15 to
afford aryl benzyl ketone 19. This material underwent cycli-
zation upon workup to yield a mixture of acid 20 and its
corresponding methyl ester (not shown). Saponification of
this mixture provided the carboxylic acid 20 in 70% overall
yield. The X-ray structure of this compound is shown in
Figure 2.5 Treatment of 20 with oxalyl chloride in the pre-
sence of a base then gave phenolic δ-lactone 21 (Scheme 3).
Next, we explored the installation of a side chain corre-
sponding to C24-C30 of the natural product. The solution
we eventually found is remarkable for its simplicity. The
achiraltetracyclic lactone 21, which functionslike anactive
ester, couldbelinked with the potassiumenolateofmethyl-
1-propenyl ketone 22 to furnish the 1,3-diketone 23, which
mostly exists in its enolized form. It should be noted that,
due to hindered rotation around the biaryl bond, this
compound is chiral and is formed as a racemate. An
analogous transformation could be carried out with the
enolate of the enantiomerically enriched epoxy ketone 24,
whose synthesis is also shown in Scheme 4.10 It involves the
oxidation of the known sorbitol monoepoxide 26 to 27
followed by the addition of methyl lithium and another
oxidation. Condensation of the potassium enolate of
ketone 24 with 21 gave aryl naphthol 25 as an inseparable
1:1 mixture of diastereomers with respect to the biaryl axis
(Scheme 4). Apparently, the stereogenic centers of the
epoxide moiety are too far away to effect dynamic kinetic
resolution and induce diastereoselectivity in the ring-open-
ing reaction.11
With 23 and 25 in hand, we began to explore the
projected key cyclizations toward the core of A-74528 (1)
(Scheme 5). Thus far, our efforts have been met with
limited success. Nevertheless, some interesting results have
been obtained. For instance, treatment of the diketone
(6) (a) Danishefsky, S. J.; Dushin, R. G. J. Am. Chem. Soc. 1992, 114,
655–659. (b) Takahashi, S.; Kamisuki, S.; Kobayashi, S.; Sugawara, F.
€
Tetrahedron 2004, 60, 5695–5700. (c) Altenmoller, M.; Podlech, J.;
Fenske, D. Eur. J. Org. Chem. 2006, 1678–1684.
(7) For sterically hindered biaryl couplings, see: Tang, T.; Capacci,
A. G.; Wei, X.; Li, W.; White, A.; Patel, N. D.; Savoie, J.; Goa, J. J.;
Rodriguez, S.; Qu, B.; Haddad, N.; Lu, B. Z.; Krishnamurthy, D.; Yee,
N. K.; Senayake, C. H. Angew. Chem., Int. Ed. 2010, 49, 5879–5883 and
references therein.
(8) (a) Chiarello, J.; Joullie, M. M. Tetrahedron 1988, 44, 41–48. (b)
Pulgarin, C.; Gunzinger, J.; Tabacchi, R. Helv. Chim. Acta 1985, 68,
1948–1951.
(10) Anson, C. E.; Dave, G.; Stephenson, G. R. Tetrahedron 2000, 56,
2273–2281.
(11) (a) Bringmann, G.; Vitt, D. J. Org. Chem. 1995, 60, 7674–7681.
(b) Bringmann, G.; Price Mortimer, A. J.; Keller, P. A.; Gresser, M. J.;
Garner, J.; Breuning, M. Angew. Chem. 2005, 117, 5518–5563. Angew.
Chem., Int. Ed. 2005, 44, 5384–5427.
(9) Williams, R. M.; Ehrlich, P. P.; Zhai, W.; Hendrix, J. J. Org.
Chem. 1987, 52, 2615–2617.
1388
Org. Lett., Vol. 13, No. 6, 2011

Scheme 4. Synthesis of Epoxy Ketone 24 and Installment of the
Enone
Scheme 5. Unexpected Cyclizations
prepared using a challenging Suzuki coupling and biomi-
metic condensation chemistry, respectively. Efforts to
convert these intermediates into our target molecule via
biomimetic cyclization cascades are currently underway in
our laboratories. Alternative cyclization modes shown in
Scheme 1 are also being explored.
23 with a weak base at high temperature provided the
γ-pyrone 29 as a mixture of two separable diastereomers.
These compounds are presumably formed via O-attack
of the enolate of the β- diketone onto the Michael system
of the side chain. Upon stirring of the diketone 23 in a
DMF/THF mixture in the presence of TBAF and air, a
complex structure was isolated, which was identified using
2D-spectroscopy data (COSY, NOESY, HMBC, HSQC)
as 30. This compound is presumably formed via an air
oxidation of 23 to the corresponding naphthoquinone,
followed by a Michael-Michael-aldol cascade (see Sup-
porting Information).3
Acknowledgment. This work was supported by the
Stiftung der Deutschen Wirtschaft (graduate fellowship
to A.H.) We thank Corinna Jansen (undergraduate re-
searcher, LMU) and Dominik Hager (University of Mu-
nich, LMU) for assistance with preparing intermediates
and elucidating spectra.
Supporting Information Available. Experimental de-
tails, spectroscopic and analytical data for all new com-
pounds (including X-ray data for 11, 13, and 20). This
material is available free of charge via the Internet at
http://pubs.acs.org.
In summary, we have synthesized a range of highly sub-
stituted biaryl compounds that might serve as valuable inter-
mediates in a total synthesis of A-74528. They were
Org. Lett., Vol. 13, No. 6, 2011
1389