Total synthesis of siphonarin B and dihydrosiphonarin B.

Article abstract of DOI:10.1021/ol017082w

The spirocyclic core of the siphonarins was constructed by a directed cyclization of a linear triketone, prepared using a Sn(II)-mediated aldol coupling and Swern oxidation at C9 and C13. To circumvent a facile retro-Claisen pathway generating a baconipyr

Full text of DOI:10.1021/ol017082w

ORGANIC
LETTERS
2002
Vol. 4, No. 3
391-394
Total Synthesis of Siphonarin B and
Dihydrosiphonarin B
Ian Paterson,* David Yu-Kai Chen, and Alison S. Franklin
UniVersity Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW, UK
ip100@cus.cam.ac.uk
Received November 21, 2001
ABSTRACT
The spirocyclic core of the siphonarins was constructed by a directed cyclization of a linear triketone, prepared using a Sn(II)-mediated aldol
coupling and Swern oxidation at C9 and C13. To circumvent a facile retro-Claisen pathway generating a baconipyrone-type ester, a Ni(II)/
Cr(II)-mediated coupling reaction with vinyl iodide was used to complete the first synthesis of siphonarin B and dihydrosiphonarin B. A stable
isomeric spiroacetal was also prepared which could not be equilibrated to the siphonarin skeleton.
Siphonarins A (1) and B (2) are unusual γ-pyrone polypro-
pionates,1-3 containing a characteristic spiroacetal ring
system, which were first isolated by Faulkner and Ireland
and their co-workers2a from the marine molluscs, Siphonaria
zelandica and S. atra, collected on the coast of New South
Wales, Australia, while the C3 dihydro congeners 3 and 4
were obtained from a siphonariid collection made in Hawaii.
Their highly substituted spiro-bis-acetal core is proposed to
be determined by the oxidation state of the carbons and the
configuration of the hydroxyl- and methyl-bearing stereo-
centers in an unstable linear polyketide-derived precursor,
such as 5, that participates in a thermodynamically driven
cyclization cascade.3,4
As part of studies on these and related siphonariid-derived
polypropionates,2c,5,6 we have previously synthesized ba-
conipyrone C (6),5a,7 which lacks a normal contiguous carbon
skeleton. By controlling the cyclization mode of an open-
chain precursor, we now report the first total synthesis of
siphonarin B (2) and dihydrosiphonarin B (4), as well as
the formation of a stable isomeric spiroacetal ring system,
along with experimental support for the occurrence of a retro-
Claisen reaction leading to baconipyrone-type structures
under mild conditions.
As outlined in Scheme 1, unravelling of the spirocyclic
core of siphonarin B (2) suggested the triketones 7 (C1-
C21) and 8 (C3-C21) as protected acyclic precursors. For
synthetic purposes, the introduction of the elaborate oxy-
(1) For reviews, see: (a) Davies-Coleman, M. T.; Garson, M. J. Nat.
Prod. Rep. 1998, 15, 477. (b) O’Hagan, D. Nat. Prod. Rep. 1989, 6, 205.
(2) (a) Hochlowski, J. E.; Coll, J. C.; Faulkner, D. J.; Biskupiak, J. E.;
Ireland, C. M.; Zheng, Q.; He, C.; Clardy, J. J. Am. Chem. Soc. 1984, 106,
6748. (b) Garson, M. J.; Jones, D. J.; Small, C. J.; Liang, J.; Clardy, J.
Tetrahedron Lett. 1994, 35, 6921. (c) Paterson, I.; Franklin, A. S.
Tetrahedron Lett. 1994, 35, 6925.
(3) For modeling studies supporting this proposal, see: Garson, M. J.;
Goodman, J. M.; Paterson, I. Tetrahedron Lett. 1994, 35, 6929.
(4) An alternative cyclization mode is available to 8-epi-5 to produce
the trioxaadamantyl ring system of caloundrin B (28, Scheme 3). Blanch-
field, J. T.; Brecknell, D. J.; Brereton, I. M.; Garson, M. J.; Jones, D. D.
Aust. J. Chem. 1994, 47, 2255.
(5) (a) Paterson, I.; Chen, D. Y.-K.; Acen˜a, J. L.; Franklin, A. S. Org.
Lett. 2000, 2, 1513. (b) Paterson, I.; Perkins, M. V. Tetrahedron 1996, 52,
1811. (c) Paterson, I.; Perkins, M. V. J. Am. Chem. Soc. 1993, 115, 1608.
(6) For recent synthetic work on other siphonariid-derived compounds,
see: (a) Perkins, M. V.; Sampson, R. A. Org. Lett. 2001, 3, 123. (b) Arjona,
O.; Menchaca, R.; Plumet, J. Org. Lett. 2001, 3, 107. (c) DeBrabander, J.;
Oppolzer, W. Tetrahedron 1997, 53, 9169. (d) Yamamura, S.; Nishiyama,
S. Bull. Chem. Soc. Jpn. 1997, 70, 2025. (e) Hoffmann, R. W.; Dahmann,
G. Liebigs Ann. Chem. 1994, 837. (f) Andersen, M. W.; Hildebrandt, B.;
Dahmann, G.; Hoffmann, R. W. Chem. Ber. 1991, 124, 2127.
(7) Manker, D. C.; Faulkner, D. J.; Stout, T. J.; Clardy, J. J. Org. Chem.
1989, 54, 5371.
10.1021/ol017082w CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/05/2002

initiate a cyclization cascade to deliver the spiro-bis-acetal
ring system.
The preparation of ketone 9 (Scheme 2) began with an
asymmetric aldol reaction between 3-pentanone and (E)-2-
methyl-2-pentenal using (-)-Ipc2BOTf.8 Narasaka-type re-
duction9 of the resulting adduct 11 (nBu2BOMe; LiBH4) gave
1,3-syn diol 12 (52%, g97:3 ds, 90% ee). Following silyl
protection, hydroboration10 of the allylic ether with thexyl-
borane (90:10 ds), and then Dess-Martin oxidation, gave
ketone 9 (67%). The aldehyde component 10 was obtained
in 89% yield from the diol 13,2c,5a having the γ-pyrone ring
already installed, by a sequence of bis-TES protection,
selective cleavage, and Dess-Martin oxidation.
The key aldol coupling between 9 and 10 was best
performed using Sn(OTf)2/Et3N,11 leading to a mixture of
adducts 14 (94%; ca. 60:40 ds in favor of the 6,8-syn-8,9-
syn isomer). After separation of the major syn isomer,
selective cleavage of the TES ether, followed by careful
Swern oxidation to preserve the C8 configuration,5b gave the
triketone 7 (74%). Exposure to buffered HF‚py deprotection
then induced cyclization. However, rather than the required
spirocyclization, hemiacetal formation between the 3-OH and
C7 ketone gave 15 (69%). Oxidative removal of the PMB
ether (DDQ) then led to the spiroacetal 16 (56%) engaging
the 11-OH, accompanied by epimerization at C8. Notably,
this acetal ring system (assigned by extensive NMR analysis
and diagnostic NOE correlations) is stabilized by a double
anomeric effect, as well as having all but one of the alkyl
substituents in equatorial positions. Despite numerous at-
tempts, using a variety of acidic conditions, spiroacetal 16
could not be isomerized to generate 3-epi-dihydrosiphonarin
B (17). After succumbing to this alternative thermodynami-
cally favored cyclization mode, a revised strategy using the
modified acyclic precursor 8 shown in Scheme 1 was
adopted. This now employed orthogonal protection, with
DEIPS and benzyl ethers at C5 and C3, respectively, and
required the late introduction of the terminal ethyl group.
As shown in Scheme 3, assembly of 8 was initiated by
preparation of ketone 18 and aldehyde 19, having a labile
TMS ether at C13, enabling later discrimination from the
C5 DEIPS ether. Here, a boron-mediated aldol reaction ((c-
Hex)2BCl, Et3N) between (R)-2012 and propionaldehyde, with
in situ reduction (LiBH4),5b gave 1,3-diol 21 cleanly (86%;
>95:5 ds). Bis-DEIPS protection, followed by hydrolysis of
the less hindered silyl ether, released the 7-OH, which was
oxidized to give ketone 18 (49%). Similarly, bis-TMS
protection of the previously described diol 13,5a followed
by selective cleavage of the primary TMS ether (K2CO3,
MeOH) and buffered Dess-Martin oxidation, gave the
genation and methylation pattern, together with avoiding
alternative modes of cyclization, were challenging and
required the careful selection of protecting groups and
oxidation conditions throughout. The readily epimerizable
C8 and C14 stereocenters were also of concern.
Inspired by our earlier denticulatin synthesis,5b the 7,9,13-
triketone 7, incorporating a silylene-protected 3,5-diol and
a PMB-protected 11-OH, was selected initially as a potential
cyclization precursor. We envisaged a C8-C9 aldol coupling
between ketone 9 and aldehyde 10, followed by oxidation
of the 9-OH and 13-OH and then release of the 5-OH to
Scheme 1
(8) Paterson, I.; Goodman, J. M.; Lister, M. A.; Schumann, R. C.;
McClure, C. K.; Norcross, R. D. Tetrahedron 1990, 46, 4663.
(9) Narasaka, K.; Pai, F. C. Tetrahedron 1984, 40, 2233.
(10) Still, W. C.; Barrish, J. C. J. Am. Chem. Soc. 1983, 105, 2487.
(11) (a) Mukaiyama, T.; Iwasawa, N.; Stevens, R. W.; Haga, T.
Tetrahedron 1984, 40, 1381. (b) Paterson, I.; Tillyer, R. D. Tetrahedron
Lett. 1992, 33, 4233.
(12) (a) Paterson, I.; Goodman, J. M; Isaka, M. Tetrahedron Lett. 1989,
30, 7121. (b) Paterson, I.; Norcross, R. D.; Ward, R. A.; Romea, P.; Lister,
M. A. J. Am. Chem. Soc. 1994, 116, 11287.
(13) The configuration was determined by NMR analysis of the derived
PMP acetal obtained on exposure of 22 to anhydrous DDQ.
392
Org. Lett., Vol. 4, No. 3, 2002

Scheme 2
i
d
a (-)-Ipc2BOTf, Pr2NEt, CH2Cl2. b nBu2BOMe, THF/MeOH; LiBH4. c tBu2Si(OTf)2, 2,6-lutidine, CH2Cl2. Thexylborane, THF; H2O2,
e
f
g
h
i
NaOH. Dess-Martin periodinane, CH2Cl2. TESOTf, 2,6-lutidine, CH2Cl2. AcOH/THF/H2O. Sn(OTf)2, Et3N, CH2Cl2. PPTS, MeOH/
CH2Cl2. (COCl)2, DMSO, CH2Cl2; Et3N. HF‚py/py, THF. DDQ, CH2Cl2/pH 7 buffer.
j
k
l
γ-pyrone aldehyde 19 (98%). This was immediately sub-
jected to reaction with the Sn(II) enolate11 derived from
ketone 18 to give a mixture of aldol adducts 22 (92%; ca.
73:27 ds in favor of the 6,8-syn-8,9-syn isomer13). Selective
deprotection of the TMS ether (PPTS, MeOH/CH2Cl2),
followed by double Swern oxidation, then gave the sensitive
triketone 8 (and its C8 epimer, ca. 2.7:1; 92%). In this case,
desilylation (HF‚py/py) led to the formation of the required
six-membered hemiacetal ring 23 (55%), having all alkyl
substituents equatorially oriented (where an NOE correlation
between H6 and H8 indicated that the major C8 configuration
had been preserved).
With hemiacetal 23 in hand, attention was now focused
on generating the spirocyclic core of the siphonarins.
However, exposure of the delicate hemiacetal 23 to mildly
acidic (e.g., extensive SiO2 chromatography) or basic condi-
tions resulted in a facile retro-Claisen reaction, producing
the baconipyrone-type ester 24 (Scheme 4), which had a
diagnostic resonance for H5 (5.35 ppm, dd, J ) 8.3, 3.4 Hz).
A similar rearrangement has been proposed by Faulkner7 for
the biosynthetic interconversion of the siphonarins and
baconipyrones. The present finding suggests that this could
easily occur on chromatographic purification, lending support
Scheme 3
b
c
d
a (c-Hex)2BCl, Et3N, Et2O, EtCHO; LiBH4. DEIPSCl, Imidazole, DMF. PPTS, MeOH/CH2Cl2. Dess-Martin periodinane, CH2Cl2.
eTMSOTf, 2,6-lutidine, CH2Cl2. fK2CO3, MeOH. gDess-Martin periodinane, py, CH2Cl2. hSn(OTf)2, Et3N, CH2Cl2. iPPTS, MeOH/CH2Cl2.
j(COCl)2, DMSO, CH2Cl2; Et3N. HF‚py/py, THF. Pd/C, H2, EtOH. mNiCl2/CrCl2 (5% NiCl2), H2CdCHI, DMF.
k
l
Org. Lett., Vol. 4, No. 3, 2002
393

Thus debenzylation of hemiacetal 23 under controlled
conditions (H2, Pd/C, EtOH, 30 min) with retention of the
PMB ether (Scheme 3), followed by Swern oxidation of the
resulting primary alcohol (the 5-OH is internally protected
here as a hemiacetal),5c gave a labile aldehyde that was
immediately subjected to Kishi-Nozaki15 coupling condi-
tions (CrCl2, 5% NiCl2, DMF) with vinyl iodide, leading to
an ca. 2.5:1 mixture of allylic alcohols 26 in 84% yield.
Notably, attempted addition of more common organometallic
reagents (EtMgBr, EtLi. Et2CuLi, etc.) led only to decom-
position by the retro-Claisen pathway, emphasizing the
mildness of the Ni(II)/Cr(II)-mediated protocol.
Scheme 4
A further Swern oxidation performed on 26 then gave the
corresponding enone, which was submitted to catalytic
hydrogenation (H2, Pd/C, EtOH, 16 h) to provide the ethyl
group with concomitant removal of the PMB ether. Gratify-
ingly, this final step also triggered the desired spirocyclization
through hemiacetalization between the 9-OH and the C13
ketone in 27, leading to isolation of (+)-siphonarin B (2).16
The spectroscopic data (1H, 13C, IR, HRMS) and specific
rotation, [R]20 +10.5 (c 0.12, CHCl3), were in excellent
D
agreement with that reported2a for the natural material ([R]20
D
+13.2 (c ) 0.014, CHCl3) and by NMR comparison with
an authentic sample, thus confirming the absolute configura-
tion.2b,c Notably, we did not detect any formation of
caloundrin B (28), an isomeric siphonariid-derived compound
that might have been produced by an alternative cyclization
mode via 8-epi-27.4 Similarly, dihydrosiphonarin B (4) could
also be obtained by catalytic hydrogenation of the major
epimer at C3 in 26 and this had spectroscopic data in accord
with that reported.2a,16
In conclusion, the first total synthesis of the unusual marine
polypropionates siphonarin B (2) and dihydrosiphonarin B
(4) has been accomplished.17 The successful generation of
the highly substituted spiro-bis-acetal core of the siphonarins
from the acyclic triketone precursor 8 required the careful
orchestration of delicate intermediates. In contrast, the more
elaborate precursor 7 was easily converted into a particularly
stable spiroacetal ring system 16 that resisted isomerization
into the siphonarin skeleton. Conceivably, this novel ring
system might even be found in siphonariid-derived extracts.4
In general, these findings lend support to the proposal that
such siphonariid-derived compounds are thermodynamic, i.e.,
non-enzymic, cyclization products of unstable acyclic polypro-
pionate metabolites.1a,3,4,5b,14
to the suggestion by Garson that the baconipyrones may be
isolation artifacts.1a,14
While hemiacetal 23 resisted all attempts at further
cyclization due to this competing retro-Claisen pathway,
hydrogenolysis of the benzyl and PMB ethers (H2, Pd/C,
EtOH, 48 h) led to isolation of the desired thermodynamically
favorable spiro-bis-acetal core 25 (32%), where all the alkyl
substituents are equatorially oriented with anomeric stabiliza-
tion at the C9 and C13 acetal centers. Recognizing that mild
reaction conditions and workup procedures (with minimal
chromatography) were crucial for the remainder of the
synthesis, this observation suggested that reductive removal
of the PMB ether might be used in the last step as a
convenient trigger for the spirocyclization.
Acknowledgment. We thank the EPSRC, the Cambridge
Commonwealth Trust, the Sim Foundation, King’s and
Corpus Christi Colleges, Merck, and Pfizer for support. We
are grateful to Dr. Jose´ Luis Acen˜a (Cambridge) and Dr.
Mary J. Garson (Queensland) for helpful discussions and to
Dr Garson for kindly providing us with an authentic sample
of siphonarin B.
(14) A reexamination of S. baconi, as reported by Garson (ref 1a), failed
to detect any baconipyrones, such that they may conceivably be artifacts
of the isolation process rather than biosynthetically related metabolites to
the siphonarins. See also: Brecknell, D. J.; Collett, L. A.; Davies-Coleman,
M. T.; Garson, M. J.; Jones, D. D. Tetrahedron 2000, 56, 2497.
(15) (a) Takai, K.; Taghashira, M.; Kuroda, T.; Oshima, K.; Utimoto,
K.; Nozaki, H. J. Am. Chem. Soc. 1986, 108, 6048. (b) Jin, H.; Uenishi,
J.-I.; Christ, W. J.; Kishi, Y. J. Am. Chem. Soc. 1986, 108, 5644.
(16) Due to the small quantity of dihydrosiphonarin prepared, an accurate
specific rotation could not be measured.
(17) Siphonarin B (2) was obtained in 25 steps (0.9% overall yield) from
(S)-3-(benzyloxy)-2-methylpropanal (ref 5a). The low yield obtained for
the final step is due to competing decomposition encountered under the
prolonged reaction time (16 h) needed to remove the PMB ether.
Supporting Information Available: Tabulated NMR and
spectroscopic data for compounds 2, 4, and 16, along with
spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL017082W
394
Org. Lett., Vol. 4, No. 3, 2002