Studies on the synthesis of pectenotoxin II: Synthesis of a C(11)-C(26) fragment precursor via [3 + 2]-annulation reactions of chiral allylsilanes

Article abstract of DOI:10.1021/ol0160250

(formula presented) A synthesis of tetracycle 2 corresponding to the C(11)-C(26) fragment of pectenotoxin II is described. The synthesis features two highly stereoselective [3 + 2]-annulation reactions of chiral allylsilanes, generated via allylboration o

Full text of DOI:10.1021/ol0160250

ORGANIC
LETTERS
2001
Vol. 3, No. 12
1949-1952
Studies on the Synthesis of
Pectenotoxin II: Synthesis of a
C(11)−C(26) Fragment Precursor via
[3 + 2]-Annulation Reactions of Chiral
Allylsilanes
Glenn C. Micalizio1 and William R. Roush*
Department of Chemistry, UniVersity of Michigan, Ann Arbor, Michigan 48109
roush@umich.edu
Received April 24, 2001
ABSTRACT
A synthesis of tetracycle 2 corresponding to the C(11)−C(26) fragment of pectenotoxin II is described. The synthesis features two highly
stereoselective [3 + 2]-annulation reactions of chiral allylsilanes, generated via allylboration of aldehydes with the chiral γ-silylallylborane 4
or the γ-silylallylboronate 19, for construction of the highly substituted C and E rings.
The pectenotoxins are a family of highly cytotoxic polyether
macrolide toxins that are active against human lung (A-549),
colon (HT-29), and breast (MCF-7) cancer cell lines.2,3
Pectenotoxin II (1, Scheme 1), produced by the dinoflagel-
lates Dinophysis fortii and D. accuminata, has also shown
selective cytotoxicity against several cell lines representing
ovarian, renal, lung, colon, CNS, melanoma, and breast
cancer, with differences in LC50 values between sensitive
and resistant cell lines of 100-fold or more.3 Pectenotoxin II
interacts with the actin cytoskeleton at a unique site and could
become an important research tool in the study of basic
cellular processes.4
enotoxin II represents a formidable synthetic challenge, as
the structure contains two spiroketals, three tertiary ethers,
three substituted tetrahydrofurans, and 19 stereocenters
embedded within a 40-carbon chain. The exquisitely complex
Scheme 1. Retrosynthetic Analysis
Although the pectenotoxins display an array of interesting
and potentially significant biological properties, only one
study on their synthesis has been reported to date.5 Pect-
(1) Holder of the 1999-2000 ACS Division of Organic Chemistry
Graduate Fellowship sponsored by Eli Lilly.
(2) Yasumoto, T.; Murata, M.; Oshima, Y.; Sano, M.; Matsumoto, G.
K.; Clardy, J. Tetrahedron 1985, 41, 1019.
(3) Jung, J. H.; Sim, C. J.; Lee, C.-O. J. Nat. Prod. 1995, 58, 1722.
(4) Spector, I.; Braet, F.; Shochet, N. R.; Bubb, M. R. Microscop. Res.
Technol. 1999, 47, 18.
10.1021/ol0160250 CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/15/2001

structure of 1 coupled with its interesting biological properties
have prompted us to initiate studies targeting the total
synthesis of this molecule. We report herein a brief and
highly stereocontrolled synthesis of tetracycle 2, correspond-
ing to the C(11)-C(26) C-D-E fragment of the natural
product, by a route that features our convergent three-
component coupling sequence for tetrahydrofuran synthesis
via chiral allylsilane intermediates.6
We recently reported that allylsilanes of general structure
6, which are readily prepared by allylboration of aldehydes
with either chiral allylborane 47 or our first-generation tartrate
ester modified γ-silylallylboronates,8 undergo highly stereo-
selective [3 + 2] annulation reactions with aldehydes to give
2,3,5-trisubstituted tetrahydrofurans.6,9 When the reaction is
performed by using BF3‚Et2O as the (nonchelating) Lewis
acid, 2,5-cis-tetrahydrofurans 7 are prepared with at least
12:1 and most often with g20:1 selectivity. On the other
hand, reactions that are performed under chelate control using
SnCl4 as the promoter provide 2,5-trans-substituted tetrahy-
drofurans (e.g., 9) with g20:1 selectivity. The chelate-
controlled conditions also permit the synthesis of tetrahy-
drofurans with quaternary centers, as illustrated by the
synthesis of 10 in Scheme 2.6
is adjacent to the C(14)-ketone, it seemed conceivable that
the natural configuration at this center could be established
at an appropriate point in the synthetic sequence by a base-
promoted epimerization reaction. This permitted us to
contemplate the synthesis of the C ring unit of 2, with
unnatural C(15) stereochemistry, via a second [3 + 2]-
annulation reaction involving 4, 5, and the aldehyde gener-
ated from deprotection and oxidation of the C(16)-OTBS
group of 3. While we have not yet demonstrated that a C(15)
epimerization sequence can be accomplished, we have
developed and report herein a remarkably brief synthesis of
tetracycle 2 that serves to define the utility of the [3 +
2]-annulation sequence for tetrahydrofuran synthesis in a
structurally complex context.
Our synthesis of aldehyde 3 begins with the known
geraniol epoxide 11 (g92% ee) (Scheme 3).10 Reduction of
the epoxy-alcohol (Red-Al, THF)11 followed by treatment
of the 1,3-diol with benzaldehyde (PPTS, PhH, reflux)
provided the corresponding benzylidene acetal, which after
reductive opening with DIBAL-H (CH2Cl2, -78 to 23 °C)12
and protection of the primary hydroxyl group (TBSCl, Et3N,
DMAP, CH2Cl2) afforded the tert-butyldimethylsilyl ether
12 in an overall yield of 71% from 11. The olefin was then
oxidatively cleaved by a two-step sequence ((i) OsO4, NMO,
acetone, pH 7 buffer (92%); (ii) Pb(OAc)4, EtOAc) to give
the C(21) aldehyde 3 which was used in subsequent
chemistry without purification.
Scheme 2
Scheme 3
Chiral allylsilane 13, required for construction of the E
ring, was synthesized by the double asymmetric silylallyl-
boration of aldehyde 3 with the γ-silylallylborane 4 (Scheme
4).7 This reaction provided the desired anti-â-hydroxyallyl-
silane as an inseparable 9-14:1 mixture of diastereomers
(77% yield), which was subsequently protected as the
corresponding triethylsilyl ether 13 (TESCl, Et3N, DMAP,
CH2Cl2, 93%). The yield of allylsilane 13 was 66% for the
four-step sequence from olefin 12. The SnCl4-promoted [3
+ 2] annulation of 13 and methyl pyruvate (5) then afforded
the tetrasubstituted tetrahydrofuran 15 in 66-75% yield
(>20:1 ds) accompanied by a small amount of the allylation
It was readily apparent that the [3 + 2]-annulation
sequence is ideally suited for the synthesis of the E ring of
pectenotoxin II via the stepwise three-component coupling
of aldehyde 3, allylborane 4, and methyl pyruvate (5).
However, we have not yet learned how to effect a nonchelate
controlled [3 + 2]-annulation reaction of allylsilanes 6 and
ketones, which would be required for the direct introduction
of the 2,5-cis stereochemistry of the C ring of pectenotoxin
II. However, recognizing that C(15) of the natural product
(5) Amano, S.; Fujiwara, K.; Murai, A. Synlett 1997, 1300.
(6) Micalizio, G. C.; Roush, W. R. Org. Lett. 2000, 2, 461.
(7) Roush, W. R.; Pinchuk, A. N.; Micalizio, G. C. Tetrahedron Lett.
2000, 41, 9413.
(8) Roush, W. R.; Grover, P. T. Tetrahedron 1992, 48, 1981.
(9) For a review of [3 + 2]-annulation reactions of allylsilanes: Masse,
C. E.; Panek, J. S. Chem. ReV. 1995, 95, 1293.
(10) Katsuki, T.; Martin, V. S. Org. React. 1996, 48, 1.
(11) Gao, Y.; Sharpless, K. B. J. Org. Chem. 1988, 53, 4081.
(12) Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K. Chem. Lett. 1983,
1593.
1950
Org. Lett., Vol. 3, No. 12, 2001

Scheme 4
Scheme 5
product 16 (11%). The stereochemistry within the tetrahy-
drofuran ring of 15 was assigned by 1H NOE studies and is
consistent with the first step of the [3 + 2]-annulation
reaction proceeding by way of the syn-synclinal transition
state 14.13 The stereochemistry of the tertiary alcohol center
in 16 is assumed to be the same as that of the quaternary
center in 15, but this has not been assigned rigorously.
Reduction of the methyl ester unit of 15 (DIBAL-H, THF
0 °C) followed by benzylation of the primary alcohol (NaH,
BnBr, tetrabutylammonium iodide (TBAI), THF, reflux)
afforded 17 in 83% yield (Scheme 5). Global desilylation
of 17 by using a modified Hudrlik protiodesilylation
protocol14,15 (5% KO-t-Bu, DMSO, H2O, TBAF, 18-crown-
6, 85 °C) afforded the C(16),C(21)-diol in 95% yield. The
diol was then oxidized via the standard Swern protocol16 to
give the keto aldehyde 18, which was used in the subsequent
step without purification. Surprisingly, in initial attempts to
γ-silylallylate 18 using the allylborane 4, the aldehyde
underwent reduction and allylation products were not
obtained. Consequently 18 was treated with our first-
generation tartrate ester modified γ-silylallylboronate (R,R)-
19 (toluene, -78 °C, 4 Å molecular sieves).8 This reaction
provided the allylsilane 20 in 80% yield from 17 with
excellent stereo- and regioselectivity; products from allylation
of the ketone carbonyl were not observed. Protection of the
hydroxyl group of 20 as a TES ether (TES-Cl, DMF,
imidazole, 70 °C, 88% yield) then set the stage for introduc-
tion of the C ring by a second [3 + 2]-annulation reaction
with methyl pyruvate. In the event, treatment of the allyl-
silane with 3 equiv of the 1:1 complex of methyl pyruvate
and SnCl4 in CH2Cl2 at -78 °C provided a 1:1 mixture of
bis-tetrahydrofuran 21 and an allylated product tentatively
assigned the stereochemistry of 22. The yield of 21 from 20
is thus 30%.
The synthesis of the C-D-E tetracyclic synthon 2 was
completed by deprotection of the TES ether (PPTS, MeOH,
77% yield) and then hydrogenolysis of the two benzyl ethers
over Pd(OH)2 on carbon in MeOH. Under these conditions,
the keto diol spontaneously cyclized to give the targeted
tetracycle 2 in 63% overall yield from 21. The stereochem-
1
istry of 2 was confirmed by the H NOE data summarized
in Scheme 6. Noteworthy are the NOE interactions observed
Scheme 6
(13) Keck, G. E.; Savin, K. A.; Cressman, E. N. K.; Abbott, D. E. J.
Org. Chem. 1994, 59, 7889.
(14) Hudrlik, P. F.; Hudrlik, A. M.; Kulkarni, A. K. J. Am. Chem. Soc.
1982, 104, 6809.
(15) Hudrlik, P. F.; Holmes, P. E.; Hudrlik, A. M. Tetrahedron Lett.
1988, 29, 6395.
(16) Tidwell, T. T. Org. React. 1990, 39, 297.
Org. Lett., Vol. 3, No. 12, 2001
1951

between the C(11)-methoxycarbonyl group and the dimeth-
ylphenylsilyl group, as well as that between the dimeth-
ylphenylsilyl unit and H(15) which collectively serve to
suppressing the allylation pathway that compromised the
efficiency of the [3 + 2]-annulation sequence leading to 21
and to defining a strategy for direct introduction of the C
ring with correct stereochemistry at C(15). These studies,
together with additional progress toward the total synthesis
of pectenotoxin II, will be reported in due course.
1
define the stereochemistry within the C ring. Similarly, H
NOE’s observed between H(22) and the C(26)-CH2OH group
confirm the stereochemistry within the trisubstituted E ring.
1
Finally, a long-range H NOE was observed between the
C(12)-Me and H(22) which is completely consistent with
the assigned structure.
Acknowledgment. Support provided by the National
Institutes of Health (GM 38436) is gratefully acknowledged.
We also thank Eli Lilly for a graduate fellowship to G.C.M.
In summary, we have developed an expeditious strategy
for the synthesis of tetracycle 2, a synthetic equivalent of
the C(11)-C(26) fragment of pectenotoxin II, by a route
that employs our recently developed three-component cou-
pling strategy for tetrahydrofuran synthesis. The synthesis
of 2 proceeds in 18 steps from geraniol epoxide (11) in 4.4%
overall yield. Studies currently in progress are focusing on
Supporting Information Available: Experimental pro-
cedures for synthesis of 13-21 and 2. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0160250
1952
Org. Lett., Vol. 3, No. 12, 2001