Enantiocontrolled construction of sistodiolynne, an unusual polyketide from the wood-decay fungus Sistrema raduloides

Article abstract of DOI:10.1039/a700186j

Sistodiolynne, an unusual polyketide isolated from the wood decay fungus Sistorema raduloides, has been constructed stereoselectively from a chiral cyclopentenol building block to verify the proposed structure.

Full text of DOI:10.1039/a700186j

View Article Online / Journal Homepage / Table of Contents for this issue
Enantiocontrolled construction of sistodiolynne, an unusual polyketide from
the wood-decay fungus Sistrema raduloides
Tsutomu Sugahara and Kunio Ogasawara*
Pharmaceutical Institute, Tohoku University, Aobayama, Sendai 980-77, Japan
Sistodiolynne, an unusual polyketide isolated from the wood
decay fungus Sistorema raduloides, has been constructed
stereoselectively from a chiral cyclopentenol building block
to verify the proposed structure.
reaction7,8 between 6 and trimethylsilylacetylene proceeded
without difficulty using dichlorobis(triphenyl-
phosphine)palladium(ii) [PdCl2(PPh3)2]9 as catalyst in the
presence of copper(i) iodide in triethylamine to give the enyne
7, [a]2D8 +70.2 (c 1.4, CHCl3), in 96% yield. To remove the two
different protecting groups, 7 was sequentially exposed to
titanium(iv) chloride2,10 and tetrabutylammonium fluoride to
give cis-1-ethynylcyclopent-1-ene-3,5-diol 9, [a]2D5 240.8 (c
0.5, CHCl3), in 70% overall yield via 8.
Sistodiolynne [(1R,3S)-4-butadiynylcyclopent-4-ene-1,3-diol]
1 is a metabolite of the wood rot decay fungus Sistorema
raduloides (P. Karst) Donk which was recovered from indoor
air of a museum in San Juan, Puerto Rico.1 The compound 1
proved to be extremely air-sensitive and only stable in solution.
It has been shown by labelling experiments that it arises from
five acetate units with loss of a methyl carbon from one of the
acetate units during the biosynthesis. Its structure was deter-
mined by spectroscopic methods, mainly, NMR, and the
absolute configuration was proposed by circular dichroism of
the di(4-bromobenzoate) of the perhydro derivative of the
natural product. In relation to our recent synthesis of chiral
cyclopentenone building blocks,2,3 we were interested in
converting our chiral cyclopentenol block 2 into sistodiolynne 1
so as to establish an enantiocontrolled synthesis and to confirm
the proposed structure. Here we report the first construction of
sistodiolynne 1 and the all-cis di(4-bromobenzoate) (2)-12 of
the perhydro derivative of 1, which verified the proposed
structure for the natural product (Scheme 1).
Very fortunately, a mild two-step construction of a 1,3-diyne
functionality from a terminal acetylene exists.11 To apply this
procedure, 9 was first coupled with cis-1,2-dichloroethylene in
the presence of tetrakis(triphenylphosphine)palladium(0)12 in
butylamine to give the dienynediol 10, accompanied by a trace
of inseparable phenylphosphine impurities, which was isolated
in the pure state as the di(tert-butyldimethylsilyl) ether 11, [a]2D7
243.8 (c 0.3, CHCl3), by treating the mixture with tert-
butyldimethylsilyl chloride in the presence of imidazole. The
overall yield of 11 from 9 was 71%. Exposure of 11 to
tetrabutylammonium fluoride (5 equiv.) in THF at room
temperature11 for 10 h furnished sistodiolynne 1 as a highly
unstable oil by concurrent desilylation and dehydrochlorination.
Although the product generated could not be isolated in a pure
The optically pure tricyclic alcohol2 (+)-2, [a]2D8 +75.4 (c 1.2,
CHCl3), mp 95 °C, obtained by lipase-mediated kinetic
resolution,2 was first transformed into the (R)-cyclopente-
none2,4 4, [a]2D8 +53.2 (c 0.9, MeOH), in 72% overall yield via
the tricyclic ketone (2)-3, [a]2D6 2186.6 (c 1.0, CHCl3), by
sequential oxidation and thermolysis (Scheme 2). Although
optically active 4 may be obtained more directly from the
cyclopentenol precursor by a similar lipase-mediated resolu-
tion2 or from racemic 4 by chemical resolution,4 the present
procedure seemed more appropriate with respect to the optical
purity of the product. Exposure of 4 to iodine in the presence of
pyridine5 gave the air sensitive a-iodo enone 5, which was
immediately reduced with sodium borohydride–cerium(iii)
chloride6 to give stereoselectively cis-4-tert-butoxy-2-iodo-
cyclopent-2-enol†‡ 6, [a]2D8 +41.0 (c 0.9, CHCl3), mp
90–92.5 °C, as a single product in 92% overall yield from 4. The
stereochemistry of 6 was determined by a NOE experiment on
the benzoate of 6, which exhibited significant interaction
between the C(1) (dH 5.70) and C(4) (dH 4.54) protons.
Moreover, the stereochemistry of 6 was confirmed unambigu-
ously by converting it into cis-4-tert-butoxycyclopent-2-enol2
in 80% yield on exposure to tert-butyllithium followed by acetic
acid in tetrahydrofuran (THF) at 278 °C. A cross-coupling
O
OBut
X
i
ii
(+)-2
ButO
O
(–)-3
4 X = H
iii
5 X = I
iv
HO
X
HO
I
RO
RO
v
Cl
viii
ButO
RO
6
7 R = But, X = SiMe3
8 R = H, X = SiMe3
9 R = X = H
10 R = H
11 R = SiMe2But
vi
vii
x
OCOC6H4Br-p
HO
S
R
S
3
1
3
1
4
S
R
OCOC6H4Br-p
(–)-(1S,3S,4R)-12
HO
(1R,3S)-sistodiolynne 1
HO
Scheme 2 Reagents and conditions: i, SO3·pyridine, Me2SO, Et3N, room
temp. (94%); ii, o-dichlorobenzene, 180 °C, 2.5 h (77%); iii, I2, pyridine,
CH2Cl2; iv, NaBH4–CeCl3, MeOH, 278 °C (92% from 5); v, CH·CSiMe3,
PdCl2(PPh3)2 (6 mol%), CuI (8 mol%), Et3N, 4 h (96%); vi, TiCl4, CH2Cl2,
0 °C, 2 min; vii, Bu4NF, THF, room temp. (70% from 8); viii,
(Z)-ClHCNCHCl, Pd(PPh3)4 (5 mol%), CuI (15 mol%), BuNH2, benzene,
0.5 h; ix, ButMe2SiCl, imidazole, DMF (71% from 9); x, Bu4NF (5 equiv.),
THF, room temp., 10 h
OBut
S
3
1
R
HO
HO
(1R,3S)-sistodiolynne 1
(+)-2
Scheme 1
Chem. Commun., 1997
767

View Article Online
For 12: dH (300 MHz, CDCl3) 7.90 (2 H, d, J 8.5 Hz), 7.74 (2 H, d, J 8.5
Hz), 7.58 (2 H, d, J 8.5 Hz), 7.46 (2 H, d, J 8.5 Hz), 5.42 (1 H, m), 5.38 (1
H, m), 2.60 (1 H, dt, J 13.5, 7.5 Hz), 2.42 (1 H, ddd, J 16, 8.0, 5.0 Hz), 2.19
(1 H, br d, J 16 Hz), 2.0 (1 H, m), 1.3–1.8 (7 H, m), 0.92 (3 H, br t, J 7.0).
M+, 522.0048; C23H24O479Br2 requires M+, 522.0041.
state as it polymerised even in a solution during purification as
described,1 its 1H NMR spectrum measured directly [in
(CD3)2CO] was identical to that of the natural product.1
On the other hand, 11 was hydrogenated under medium-
pressure conditions in the presence of palladized carbon to
convert it into the mixture containing the stable perhydro
derivative of the natural product 1 as the major product. The
major product isolates as the di(4-bromobenzoate) 12, after
desilylation and 4-bromobenzoylation, had a circular dichroism
(CD) spectrum, showing the first Cotton effect (negative) at 251
nm and the second (positive) at 234 nm, which was identical to
that reported for 12 having the 1S,3S,4R configuration (negative
at 251 nm and positive at 231 nm) originated from the natural
product 1. This indicated the stereochemistry of sistodiolynne 1
to be 1R,3S as proposed.
References
1 A. K. Amegadzie, W. A. Ayer and L. Sigler, Can. J. Chem., 1995, 73,
2119.
2 T. Sugahara, Y. Kuroyanagi and K. Ogasawara, Synthesis, 1996,
1101.
3 A recent review for the chiral cyclopentenoid derivatives, see
K. Ogasawara, J. Synth. Org. Chem. Jpn., 1996, 54, 15.
4 B. M. Eschler, R. K. Haynes, M. D. Ironside, S. Kremmydas,
D. D. Riley and T. W. Hambley, J. Org. Chem., 1991, 56, 4760.
5 C. R. Johnson, P. A. Adams, M. P. Braun, C. B. W. Senanayake,
P. M. Wovkulich and M. R. Uskokovic, Tetrahedron Lett., 1992, 33,
1917.
6 A. L. Gemal and T. L. Luche, J. Am. Chem. Soc., 1981, 103, 5454.
7 K. Sonogashira, Y. Toda and N. Hagiwara, Tetrahedron Lett., 1975,
4467.
In summary, we could not isolate sistodiolynne 1 in a pure
form owing to its intrinsic instability, however, the present
synthesis verified the correctness of the proposed structure of
the natural product made by spectroscopic methods.
Footnotes
8 A more related example, see T. Kamikubo and K. Ogasawara, Chem.
Commun., 1996, 1679.
* E-mail: konol@mail.cc.tohoku.ac.jp
† Spectral (IR, 1H NMR, mass and analytical (combustion and/or high
resolution mass) data were obtained for all isolable compounds.
‡ Selected data for representative compounds. For 6: dH (300 MHz, CDCl3)
6.23 (1 H br s), 4.41 (2 H, m), 2.73 (1 H, dt, J 14, 7 Hz), 1.71 (1 H, dt, J 14,
5 Hz), 1.10 (9 H, s). For 11: dH (300 MHz, CDCl3) 6.36 (1 H, d, J 7.3 Hz),
6.12 (1 H, d, J 1.8 Hz), 5.98 (1 H, d, J 7.3 Hz), 5.05 (1 H, br s), 4.98 (1 H,
br s), 2.03 (2 H, m), 0.89 (9 H, s), 0.87 (9 H, s), 0.1 (6 H, s), 0.06 (6 H, s).
M+, 412.2017; C21H37O2Si235Cl requires M+, 412.2022. For 1: dH [300
MHz, (CD3)2CO] 6.32 (1 H, s), 4.65 (2 H, m), 4.46 (1 H, d, J 6.5 Hz), 4.17
(1 H, d, J 6.5 Hz), 3.27 (1 H, s), 2.70 (1 H, ddd, J 13.2, 7.3, 7.3 Hz), 1.45
(1 H, J 13.2, 6.0, 6.0 Hz). M+, 148.0494; C9H8O2 requires M+, 412.2022.
9 W. J. Scott, M. R. Pena, K. Sward, S. J. Stoessel and J. K. Stille, J. Org.
Chem., 1985, 50, 2302; E. J. Corey, M.-C. Kang, M. C. Desai,
A. K. Ghosh and I. N. Houpis, J. Am. Chem. Soc., 1988, 110, 649.
10 R. H. Schlessinger and R. A. Nugent, J. Am. Chem. Soc., 1982, 104,
1116.
11 A. S. Kende and C. A. Smith, J. Org. Chem., 1988, 53, 2655.
12 V. Ratovelomanana and G. Linstrumelle, Tetrahedron Lett., 1981, 22,
315.
Received in Cambridge, 8th January 1997; Com. 7/00186J
768
Chem. Commun., 1997