Total Synthesis of Filipin III

Full text of DOI:10.1021/ja971916h

12360
J. Am. Chem. Soc. 1997, 119, 12360-12361
Scheme 1
Total Synthesis of Filipin III
Timothy I. Richardson and Scott D. Rychnovsky*
Departments of Chemistry, UniVersity of
California, IrVine, California, 92697 and
UniVersity of Minnesota
Minneapolis, Minnesota 55455
ReceiVed June 11, 1997
Filipin, a polyene macrolide antibiotic, was isolated from cell
culture filtrates of Streptomyces filipinensis1 and subsequently
shown to be a mixture of four components: filipin I (4%), II
(25%), III (53%), and IV (18%).2 The structure of filipin III,
the major component of the filipin complex, was assigned in a
series of degradation studies.3 Recently, we completed the
structure determination of filipin III (1) by reporting its relative
and absolute stereochemistry.4 Filipin is structurally and
functionally distinct from both the oxopolyene macrolides5 and
the more common mycosamine-containing polyene macrolides
like amphotericin B.6 Filipin is a membrane disrupter that
selectively binds cholesterol.7 It has found widespread use as
a histochemical stain for cholesterol and has even been used to
quantitate cholesterol in cell membranes.8 Described herein is
the first total synthesis of filipin III (1).9
Scheme 2
The polyol segment of filipin was to be assembled using
cyanohydrin acetonide 4, which contained the pentane side chain
and the C1 carboxylic acid masked as an alkene. The synthesis
of cyanohydrin 4 (Scheme 1) began with the allylic alcohol 5,
which was obtained in 98% yield by Red-Al reduction of
2-octyn-1-ol. Sharpless asymmetric epoxidation of 5 provided
the expected epoxide in near quantitative yield and 99% ee.
Copper catalyzed nucleophilic opening of the epoxide with vinyl
magnesium bromide gave a 2:1 mixture of regioisomers.11 The
minor, undesired 1,2-diol was oxidatively cleaved with periodate
to facilitate purification of the desired 1,3-diol 7. A three step
protection-deprotection sequence provided mono-TBS pro-
tected diol 8. Oxidation of 8 to the aldehyde with Dess-Martin
reagent12 followed by reaction with ethyl diazoacetate catalyzed
by SnCl2 using Roskamp’s procedure13 provided â-keto ester 9
in 81% yield. The C3 R-alcohol was most conveniently
prepared by sodium borohydride reduction of the ketone, which
gave 75% of the desired alcohol 10 and 20% of the â-isomer.
Protection of the free alcohol of 10 with TMS followed by
reduction of the ester to an aldehyde and cyanohydrin acetonide
formation completed the synthesis of cyanohydrin acetonide 4.
The polyene segment 2 was prepared as outlined in Scheme
2. Protected butane triol 11 was obtained conveniently on a
large scale from L-ascorbic acid.14 The acetonide was cleaved
with CSA and MeOH, and then both alcohols were protected
as TBS ethers. Selective hydrolysis of the primary TBS with
PPTS in MeOH gave 12, and oxidation with Dess-Martin
reagent produced aldehyde 13. Aldehyde 13 was treated with
the Grignard reagent derived from 1-(tributylstannyl)-4-ethox-
ybutadiene (Wollenberg’s reagent)15 followed by mesylation and
solvolysis16 of the secondary alcohol to give the expected dienal.
Repeating this sequence provided tetraenal 14 in 65% overall
yield. Synthesis of the polyene was completed by removing
the benzoate protecting group of 14 using a three step
procedure: protection of the aldehyde as the dimethyl acetal,
cleavage of the benzoate with DIBAL-H, and then regeneration
of the aldehyde with Amberlyst acid resin in MeOH. Tetraenal
2 degraded quickly on standing under ambient light and was
best prepared and used immediately.
(1) (a) Whitfield, G. B.; Brock, T. D.; Ammann, A.; Gottlieb, D.; Carter,
H. E. J. Am. Chem. Soc. 1955, 77, 4799.
The polyol segment of filipin III was assembled as outlined
in Scheme 3. Cyanohydrin acetonide 15, the key 1,3-diol
synthon in our iterative polyol strategy, was prepared in ca. 94%
ee as previously described.16 Alkylation of cyanohydrin ac-
etonide 15 with alkyl iodide 1617 provided 17 in good yield.
Alkyl iodide 18 was formed in excellent yield from 17 under
forcing Finkelstein conditions.16 Alkylation of a second equiv
of 15 with alkyl iodide 18 gave 19, and subsequent conversion
to the iodide 20 proceeded uneventfully. Deprotonation of
cyanohydrin acetonide 4 and alkylation with iodide 20 gave
the coupled product 21 in 70-80% yield.18 Reductive decya-
nation with concurrent deprotection of the benzyl group gave
(2) Bergy, M. E.; Eble, T. E. Biochemistry 1968, 7, 653.
(3) (a) Berkoz, B.; Djerassi, C. Proc. Chem. Soc. 1959, 316. (b) Djerassi,
C.; Ishikawa, M.; Budzikiewicz, H.; Shoolery, J. N.; Johnson, L. F.
Tetrahedron Lett. 1961, 383. (c) Ceder, O.; Ryhage, R. Acta Chem. Scand.
1964, 18, 558.
(4) (a) Rychnovsky, S. D.; Richardson, T. I. Angew. Chem., Int. Ed. Engl.
1995, 34, 1227. (b) Richardson, T. I.; Rychnovsky, S. D. J. Org. Chem.
1996, 61, 4219.
(5) Rychnovsky, S. D. Chem. ReV. 1995, 95, 2021.
(6) Omura, S.; Tanaka, H. In Macrolide Antibiotics: Chemistry, Biology
and Practice; Omura, S., Ed.; Academic Press: New York, 1984; pp 351.
(7) Gale, E. F. In Macrolide Antibiotics: Chemistry, Biology, and
Practice; Omura, S., Ed.; Academic Press: New York, 1984; pp 425.
(8) Behnke, O.; Tranum-Jensen, J.; Van Deurs, B. Eur. J. Cell Biol. 1984,
35, 189.
(9) Synthetic work on polyene macrolides has been reviewed: (a) Beau,
J. M. In Recent Progress in the Chemical Synthesis of Antibiotics; Lukacs,
G., Ohno, M., Eds.; Springer: Berlin, 1990; pp 135. Also see ref 5.
(10) Tius, M. A.; Fauq, A. H. J. Org. Chem. 1983, 48, 4131. (b) Roush,
W. R.; Ando, K.; Powers, D. B.; Palkowitz, A. D.; Halterman, R. L. J. Am.
Chem. Soc. 1990, 112, 6339.
(11) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
(12) Holmquist, C. R.; Roskamp, E. J. J. Org. Chem. 1989, 54, 3258.
(13) Abushanab, E.; Vemishetti, P.; Leiby, R. W.; Singh, H. K.;
Mikkilineni, A. B.; Wu, D. C.-J.; Saibaba, R.; Panzica, R. P. J. Org. Chem.
1988, 53, 2598.
(14) Wollenberg, R. H. Tetrahedron Lett. 1978, 717.
(15) Corey, E. J.; Clark, D. A.; Goto, G.; Marfat, A.; Mioskowski, C. J.
Am. Chem. Soc. 1980, 102, 1436.
(16) Rychnovsky, S. D.; Griesgraber, G. J. Org. Chem. 1992, 57, 1559.
(17) Berlage, U.; Schmidt, J.; Peters, U.; Welzel, P. Tetrahedron Lett.
1987, 28, 3091.
(18) The addition of DMPU to this coupling reaction resulted in lower
yield due to accelerated degradation of the anion 4. The anion was stable
in THF without DMPU at -30 °C, and prolonged alkylation times (16 h)
at this temperature resulted in consistently high yields.
S0002-7863(97)01916-1 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12361
Scheme 3
polyol 22 as a single diastereomer.19 Oxidation to the aldehyde,
addition the lithium anion of ethyl diethylphosphonate, and
oxidation of the resulting secondary alcohol provided â-keto
phosphonate 23 as a 1:1 mixture of diastereomers. Stepwise
oxidation of the olefin in 23 gave the carboxylic acid 3 in
excellent yield and completed the synthesis of the polyol
segment. This highly convergent and efficient strategy provided
us with multigram quantities of the advanced intermediate 21.
Union of the polyene segment 2 and the polyol segment 3
proved difficult due to steric congestion about the C1 carboxylic
acid. After screening a variety of standard methods it was found
that this coupling could be accomplished efficiently using
Yamaguchi’s esterification protocol.20 The difficulty encoun-
tered in this apparently simple esterification reaction convinced
us to focus on a Horner-Emmons cyclization route rather than
the alternative macrolactonization. The formation of trisubsti-
tuted alkenes in macrocyclizations is uncommon.21 We are
aware of only one other ketophosphonate macrocyclization to
form a trisubstituted alkene,22 and base-induced elimination was
not an issue in that case. Formation of the macrocycle from
24 was unsuccessful under the Masamune conditions (LiCl and
DBU).23 The macrocyclization did proceed in reasonable yield
when the reaction was initiated with K2CO3 and 18-crown-6 in
warm toluene.24 As predicted, the C15 ketone of 25 was
reduced to the C15-(S) alcohol 26 with 3:1 selectivity.25 All
that remained was the deprotection of macrocycle 26.
Omura classifies filipin III as a methylpentaene macrolide,
and the C16 methyl group confers increased acid lability when
compared with other polyene macrolide antibiotics.6 Both the
acetonide and TBS protecting groups in 26 are acid labile, but
direct deprotection under acidic conditions was unsuccessful
due to competitive decomposition. The acid lability of 26 was
presumably associated with solvolysis of the C15 or C26 allylic
alcohols. Similar mixtures of degradation products were
generated by exposing 26 or natural filipin III to acidic
conditions. Alcohols protected as silyl ethers can have reduced
propensity toward elimination. Thus, the TIPS protected
derivative 27 was prepared and exposed to PPTS in warm
methanol until most of the acetonides were cleaved as judged
by TLC. These conditions also resulted in incomplete cleavage
of the TBS groups. Final deprotection of the mixture by
treatment with HF-pyridine provided synthetic filipin III.
Synthetic and natural filipin III were identical by 1H NMR and
HPLC.
Acknowledgment. Support has been provided by the NSF Presi-
dential Young Investigator Program, Pfizer Inc., the Alfred P. Sloan
Foundation, and the University of California, Irvine. We also acknowl-
edge the support of the University of Minnesota for a dissertation
fellowship to T.I.R.
(19) Rychnovsky, S. D.; Zeller, S.; Skalitzky, D. J.; Griesgraber, G. J.
Org. Chem. 1990, 55, 5550.
(20) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989.
Supporting Information Available: Complete spectral data for
compounds 2-14, 17-27, and filipin III and modeling of 25 (10 pages).
See any current masthead page for ordering and Internet access
instructions.
(21) (a) Tius, M. A.; Fauq, A. H. J. Am. Chem. Soc. 1986, 108, 1035.
(b) Tius, M. A.; Fauq, A. J. Am. Chem. Soc. 1986, 108, 6389. (c) Kodama,
M.; Shiobara, Y.; Sumitomo, H.; Fukuzumi, K.; Minami, H.; Miyamoto,
Y. Tetrahedron Lett. 1986, 27, 2157. (d) Marshall, J. A.; DeHoff, B. S.
Tetrahedron 1987, 21, 4849. (e) Marshall, J. A.; DuBay, W. J. J. Org.
Chem. 1994, 59, 1703. (f) Williams, D. R.; McGill, J. M. J. Org. Chem.
1990, 55, 3457. (g) Rychnovsky, S. D.; Khire, U. R.; Yang, G. J. Am. Chem.
Soc. 1997, 119, 2058.
JA971916H
(22) Smith, A. B.; Dorsey, B. D.; Visnick, M.; Maeda, T.; Malamas, M.
S. J. Am. Chem. Soc. 1986, 108, 3110.
(25) The major product of the reduction was predicted based on the most
accessible face of the macrocycle. The macrolide 25 was modeled using a
Monte Carlo search in Macromodel, and the minimum energy conformation
is illustrated in the Supporting Information. Ratio was determined by NMR
analysis of the C15 acetate prepared by treating 26 with Ac2O and DMAP
in THF. The major, desired diastereomer was isolated by flash chroma-
tography.
(23) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
(24) (a) Stork, G.; Nakamura, E. J. Org. Chem. 1979, 44, 4010. (b)
Nicolaou, K. C.; Seitz, S. P.; Pavia, M. R.; Petasis, N. A. J. Org. Chem.
1979, 44, 4011.