Synthesis of the core structure of the fungal metabolite benesudon: Use of oxidative decarboxylation

Article abstract of DOI:10.1021/ol052131g

(Chemical Equation Presented) The core structure (5) of the fungal metabolite benesudon (1) was synthesized, the key step being oxidative decarboxylation of acid 17.

Full text of DOI:10.1021/ol052131g

ORGANIC
LETTERS
2
005
Vol. 7, No. 25
581-5583
Synthesis of the Core Structure of the
Fungal Metabolite Benesudon: Use of
Oxidative Decarboxylation
5
Derrick L. J. Clive,* Minaruzzaman, and Haikang Yang
Chemistry Department, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G2
derrick.cliVe@ualberta.ca
Received September 3, 2005
ABSTRACT
The core structure (5) of the fungal metabolite benesudon (1) was synthesized, the key step being oxidative decarboxylation of acid 17.
1
Benesudon (1) is a fungal metabolite with a wide range of
biological properties covering antifungal, antibacterial, cy-
totoxic, phytotoxic, and nematicidal activity. The compound
represents a very rare structural type, and a literature search
ester, enol ether) are incorporated within its compact
framework. The corresponding cores of 3 and 4 share all
these features, except for the exocyclic double bond. We
describe here the first synthesis of 5, which we have prepared
as a preliminary study for the synthesis of benesudon.
2
3
based on substructure 2 retrieved only 3 (cyclogregatin)
and 4 (aigialone), besides benesudon.
4
Our plan was to attach the exocyclic methylene unit last,
so the immediate objective became the ketene acetal 6, which
we felt might be reached from 7 or 8 by a proper choice of
X (Scheme 1). To restrict double bond formation to C(3a)-
No synthetic work has been reported on any of these
substances. Apart from its rarity, the core structure 5 of
benesudon is very densely functionalized in that several
subunits (R-methylene carbonyl, ketene acetal, vinylogous
Scheme 1
(
9.
1) Thines, E.; Arendholz, W.-R.; Anke, H. J. Antibiot. 1997, 50, 13-
1
(2) A search of the Beilstein database located only 1, but use of SciFinder
Scholar retrieved 1, 3, and 4.
3) Anke, H.; Casser, I.; Schrage, M.; Steglich, W. J. Antibiot. 1988,
1, 1681-1684.
4) Vongvilai, P.; Isaka, M.; Kittakoop, P.; Srikitikulchai, P.; Kongsaeree,
P.; Thebtaranonth, Y. J. Nat. Prod. 2004, 67, 457-460.
(
4
C(7a) (see Scheme 1), the route via 8 would probably require
that, in the intended synthesis of benesudon, the group X
(
1
0.1021/ol052131g CCC: $30.25
© 2005 American Chemical Society
Published on Web 11/15/2005

should have a particular stereochemistry (which would
depend on the nature of X) relative to both the single
hydrogen that would be present at C(4) and to the hydrogen
at C(7a). On the other hand, a route via 7 might have less
demanding stereochemical requirements because the stere-
ochemical relationship between X and only a single
hydrogensthat at C(3a)smight be important. In the event,
this was the route we used in the present work, with X )
This acid is not particularly stable and should be used
promptlyspreferably within an hour of isolation. Attempts
to generate the acid by classical hydrolysis (LiOH, MeOH-
water) were unsuccessful.
With the acid in hand, we first considered the possibility
of Barton decarboxylation and trapping of the intermediate
1
0
radical with a chalcogenide, the intention being to place a
PhSe, PhS, or 2-pyridylthio group at the ring junction (see
2
CO H.
7
5 4
, X ) SePh, SPh, C H NS), so that oxidation to the
2
The acid corresponding to 7 (X ) CO H) was made, as
corresponding selenoxide or sulfoxide would lead to the
desired alkene. To our surprise, trial experiments directed
along these lines were unpromising, and so we turned our
attention to the process of oxidative decarboxylation, which
summarized in Scheme 2. Dihydropyran (9) was converted
Scheme 2
can be effected by the action of Pb(OAc)
4
in the presence
of a cuprous salt. When we first treated acid 17 with Pb-
OAc) and Cu(OAc) ‚H O in the presence of pyridine,
1
1
(
4
2
2
11a
closely following a published general procedure, the result
was again disappointing, but a very small amount of the
desired alkene 6 was indeed produced. From this point, we
were able to modify the reaction conditions so as to afford
6
in acceptable yield (78% from ester 16) (Scheme 3). In
Scheme 3
5
efficiently into the unsaturated nitrile 10 by bromination,
displacement by cyanide, and elimination of HBr, following
5
a published procedure. Base hydrolysis then gave the
5
corresponding acid 11 (97%), which was esterified (11 f
6
1
2, Me
2
SO
4
, NaHCO
3
, 81%).
our optimized procedure, Cu(OAc) ‚H O is added to a
solution of freshly prepared acid 17 in dry PhH, followed,
2
2
Bromination of ester 12, followed by addition of the
resulting dibromides 13 to a mixture of propargyl alcohol,
Å molecular sieves, and AgOCOCF , led efficiently (83%
7
after 5 min, by Pb(OAc) , which is added in portions, in the
4
4
3
overall) to the two isomeric bromides 14. Although these
could be separated, it was more convenient to process them
(
10) (a) Replacement of carboxyl by PhSe or PhS: Barton, D. H. R.;
Bridon, D.; Zard, S. Z. Heterocycles 1987, 25, 449-462. (b) Replacement
of carboxyl by thiopyridyl unit: Barton, D. H. R.; Crich, D.; Motherwell,
W. B. Tetrahedron 1985, 41, 3901-3942.
as a mixture. Radical cyclization occurred smoothly in the
8
presence of Bu
3
SnH, under standard conditions, to afford
(
11) (a) Bacha, J. D.; Kochi, J. K. Tetrahedron 1968, 24, 2215-2226.
the desired bicyclic skeleton 15 (77%). Then, ozonolysis with
reductive workup (Ph P) liberated keto ester 16 (83%), from
which the desired acid 17 was obtained (90%) by the action
(b) Sheldon, R. A.; Kochi, J. K. Org. React. 1972, 19, 279-421. (c) Kochi,
A. J.; Bacha, J. D. J. Org. Chem. 1968, 33, 2746-2754.
3
(
12) Cu(OAc)2‚H2O (56 mg, 0.28 mmol) was added to a stirred solution
of the crude acid 17 [from ester 16 (96 mg, 0.48 mmol)] in dry PhH (2.5
9
mL) (N atm), and stirring was continued for 5 min. The flask was then
of (Bu
3
Sn)
2
O in refluxing PhH for a controlled time (5 h).
2
wrapped in aluminum foil, and Pb(OAc)4 (118 mg, 0.27 mmol) was tipped
in. Stirring was continued for 30 min, and another portion of Pb(OAc)4 (55
mg, 0.13 mmol) was added, followed by PhH (1.5 mL). Stirring was again
continued for 30 min, and again a further portion of Pb(OAc)4 (88 mg,
0.20 mmol) was added, followed by PhH (1 mL) and dry DMF (0.4 mL).
The flask was fitted with a reflux condenser and flushed well with N2 for
30 min (in some experiments, the apparatus was evacuated with the house
vacuum and then filled with N2, and the process was repeated twice more).
The mixture was refluxed for 11 h (oil bath at 84 °C). The aluminum foil
was removed, and refluxing was continued for 1 h. The resulting green
solution was cooled to room temperature, evaporated to a small volume,
and applied directly to a flash chromatography column. Flash chromatog-
raphy over silica gel (1.5 × 26 cm), using 1:1 EtOAc-hexane, and then
pure EtOAc, gave 6 (52 mg, 78% over two steps) as a white solid.
(
5) Hoffmann, H. M. R.; Giesel, K.; Lies, R.; Ismail, Z. M. Synthesis
986, 548-551.
6) Faivre, V.; Lila, C.; Saroli, A.; Doutheau, A. Tetrahedron 1989, 45,
765-7782.
7) The bromination product was not characterized; we assume it is a
mixture of cis and trans dibromides.
8) A solution of Bu3SnH (0.3 M) in PhH, containing a catalytic amount
of AIBN, was added over 10 h to a stirred and heated (85 °C) solution of
4 (0.05 M) in PhH. Heating was continued for 2 h after the end of the
addition.
1
(
7
(
(
1
(
9) Salomon, C. J.; Mata, E. G.; Mascaretti, O. A. J. Org. Chem. 1994,
5
9, 7259-7266.
5582
Org. Lett., Vol. 7, No. 25, 2005

dark, and with a short (30 min) period of stirring between
each addition. Finally, a small amount of DMF11b,c is added,
the mixture thoroughly purged with N , and refluxed for 11
2
compound is a white, sharp-melting (63-64 °C) solid.13 It
is not especially sensitive and could, for example, be
chromatographed over silica gel without any noticeable
hydrolytic damage; it appeared to be more robust than the
parent acid 17, which must be processed very promptly.
h. Under these conditions (described more fully in ref 12),
the oxidative decarboxylation reliably generates alkene 6 on
a 100 mg scale, and it was then a comparatively simple
matter to attach the exocyclic methylene group.
Deprotonation of 6 (LDA, THF, -78 °C) and alkylation
with MeI gave the expected product 18, although the yield
was poor (47%) because of loss through extensive dimethy-
lation (ca. 42% yield). From that point, phenylselenation
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada and Crompton
Corp. for financial support.
(
LDA, THF, -78 °C, PhSeCl) led to selenide 19, and finally,
Supporting Information Available: Preparation of 17;
characterization data for 14, 15, 16, 17, 6, 18, 19; C NMR
oxidation (H ) gave the core of benesudon (19 f 5). The
2
O
2
13
(
13) Compound 5 had: FTIR (CH2Cl2, cast) 1595 cm-1; 1H NMR (500
MHz, CDCl3) δ 1.97-2.02 (m, 2H), 2.37 (t, J ) 6.3 Hz, 2H), 4.53 (t, J )
spectra of 5, 6, 14 (less polar isomer), 14 (more polar isomer),
and 15-19. This material is available free of charge via the
Internet at http://pubs.acs.org.
13
5
.8 Hz, 2H), 5.12 (d, J ) 2.8 Hz, 1H), 5.53 (d, J ) 2.8 Hz, 1H); C NMR
(
(
100 MHz, CDCl3) δ 15.4 9 (t), 21.5 (t), 72.0 (t), 90.4 (s), 95.8 (t), 152.8
s), 178.4 (s), 180.7 (s); exact mass m/z calcd for C8H8O3 152.04735, found
1
52.04742.
OL052131G
Org. Lett., Vol. 7, No. 25, 2005
5583