An effective extension of the polyacetate chain in the polyene macrolide antibiotic filipin III, based on chiral oxazaborolidinone-promoted asymmetric aldol reactions

Article abstract of DOI:10.1016/S0040-4039(00)01289-2

The enantioselective construction of the iterative polyacetate unit 3 in filipin III has been accomplished, essentially by an aldol strategy based on chiral oxazaborolidinone-promoted asymmetric aldol reactions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01289-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 7511–7516
An effective extension of the polyacetate chain in the
polyene macrolide antibiotic filipin III, based on chiral
oxazaborolidinone-promoted asymmetric aldol reactions
Syun-ichi Kiyooka,a,* Mostofa A. Hena,a Tomomi Yabukami,a Kazunori Muraia and
Fumitaka Gotob
aDepartment of Chemistry, Faculty of Science, Kochi University, Akebono-cho, Kochi 780-8520, Japan
bCellular Technology Institute, Otsuka Pharmaceutical Co., Ltd., Kagasuno, Kawauchi-cho,
Tokushima 771-0130, Japan
Received 30 June 2000; revised 21 July 2000; accepted 24 July 2000
Abstract
The enantioselective construction of the iterative polyacetate unit 3 in filipin III has been accomplished,
essentially by an aldol strategy based on chiral oxazaborolidinone-promoted asymmetric aldol reactions.
© 2000 Elsevier Science Ltd. All rights reserved.
A variety of approaches have been reported for the stereoselective construction of the
1,3-polyacetate moiety in polyene macrolide antibiotics.1 The access using asymmetric aldol
reactions is now limited, in spite of it being a straightforward means of stereoselectively
introducing the b-hydroxy function. It was considered that this tactic suffers from the iterative
nature required to set each stereogenic center correctly.1f By taking advantage of the iterative
nature found in 1,3-polyols, however, our aldol strategy using chiral oxazaborolidinone, (S)-1 or
(R)-1, -promoted asymmetric aldol reaction under promoter control on enantioselective acyclic
stereoselection (Fig. 1) is not necessarily the case.2 We disclose herein a versatile approach to the
1,3-polyacetate units using our aldol reaction.
Figure 1. Chiral oxazaborolidinones and facial selection
* Corresponding author. Fax: +81-88-844-8360; e-mail: kiyo@cc.kochi-u.ac.jp
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01289-2

7512
Filipin III, a polyene macrolide antibiotic which is a membrane disrupter that selectively binds
cholesterol, was isolated from cell culture filtrates of Streptomyces filipinensis.3 The structure
determination4 and the first total synthesis were achieved by the same group.5 The iterative nine
1,3-polyol unit involved in filipin III is a synthetic target suitable for the purpose of examining
the viability of our aldol strategy. The tactic using our aldol reaction repeatedly, applied to the
synthesis of an enantiopure 1,3-polyacetate unit in bryostatins, is possibly available for this
1,3-polyol unit. However, such a linear approach is not feasible in the case of so much iteration.
As shown in Scheme 1, the target polyol unit 3 was retrosynthetically planned to be divided into
two parts; segments A and B. The starting aldehyde 4, [h]2D4 −17.8 (c 1.4, CHCl3), toward the
preparation of segment A has been synthesized and reported in an enantiopure form by a
method using our asymmetric aldol reaction.6
Scheme 1. Retrosynthesis of filipin III
The synthesis of segment A is summarized in Scheme 2, which is basically realized by applying
the enantioselective aldol reaction with two different silyl nucleophiles and the following syn
selective reduction. The first aldol reaction with acetonide aldehyde 4 was carried out in the
presence of chiral oxazaborolidinone (S)-1 using silyl nucleophile 5, derived from a-unsubsti-
tuted acetate, to give the expected aldol 6 in moderate diastereoselection (75% de). In this
reaction dithiolane silyl nucleophile 7,7 developed by us and which promises quite high
enantioselection, did not work because of the excessive steric hindrance in aldehyde 4. Subjec-
tion of aldol 6 to a typical reaction sequence (DIBALH reduction and Swern oxidation), after
protection of the free hydroxy group as TBS ether, afforded the second aldehyde 8. In the
presence of the same promoter (S)-1, aldehyde 8 participated in the second aldol reaction with
disilyl nucleophile 92d,8 resulting in syn diastereoselection (92% de) to hydroxy keto ester 10. A
complete syn reduction of the keto function in 9 could, as expected, be achieved with Et2BOMe
9
and NaBH4 to give compound 11 having the four successive hydroxy functions as a 1,3-poly-

7513
Scheme 2. Enantioselective synthesis of segment A. Reaction conditions: (i) (S)-1 (BH3·THF, N-Ts-
L-valine), CH2Cl2,
−78°C, 3 h, 72%, 75% de; (ii) TBSCl, DMF, >95%; (iii) DIBALH, CH2Cl2, −78°C to rt, 84%; (iv) Swern oxid., 78%;
(v) (S)-1 (BH3·THF, N-Ts- -valine), CH2Cl2, −78°C, 6 h, 65%, 92% de; (vi) Et2BOMe, NaBH4, MeOH, >95%, >98%
L
de; (vii) CSA, Me2C(OMe)2, acetone, >95%; (viii) DIBALH, CH2Cl2, −78°C, 85%
acetate mode in excellent yield. After protection as acetonide, the following DIBALH reduction
at −78°C directly gave segment A, [h]2D5 −40.5 (c 1.26, CHCl3), as an aldehyde available for the
last condensation reaction.
The synthesis of segment B is very simple, demonstrating plainly the advantage of our aldol
strategy for constructing the 1,3-anti diol system which is obtainable only by using promoter
(S)-1 and its reverse (R)-1 twice with essentially complete selection, as depicted in Scheme 3.
Although the action of silyl nucleophile 7 on benzyl aldehyde 12 effected the desired, highly
enantioselective aldol reaction in the presence of (S)-1 to afford aldol 13 (through the si facial
selection), the dithiolane moiety in the product must be eliminated. The desulfurization
procedure was accelerated on treatment with n-Bu3SnH and AIBN10 in a quantitative yield.
After protection with TMSOTf and 2,6-lutidine, aldehyde 14 was obtained from the precursor
ester by direct DIBALH reduction. In the presence of chiral borane (R)-1, the next aldol
reaction of 14 with 7 took place in an essentially complete diastereo-controlled manner allowing
the re facial selection of the nucleophile independently of the pre-existing chiral center so as to
afford only anti isomer 15, as expected. After desilylation and acetalization, segment B, [h]D26
−2.86 (c 0.7, CHCl3), was prepared through a Weinreb amide.11

7514
Scheme 3. Enantioselective synthesis of segment B. Reaction conditions: (i) (S)-1 (BH3·THF, N-Ts-L-valine), CH2Cl2,
−78°C, 20 h, 68%, 98% ee; (ii) n-Bu3SnH, AIBN, >95%; (iii) TMSOTf, 2,6-lutidine, 85%; (iv) DIBALH, CH2Cl2,
−78°C, 72%; (v) (R)-1 (BH3·THF, N-Ts- -valine), CH2Cl2, −78°C, 20 h, 65%, >98% de; (vi) citric acid, MeOH, >95%;
(vii) CSA, Me2C(OMe)2, acetone, >95%; (viii) Me(MeO)NH·HCl, i-PrMgBr, 84%; (v) MeMgI, 92%
D
Scheme 4. Condensation of segments A and B. Reaction conditions: (i) (S)-1 (BH3·THF, N-Ts-
L-valine), C2H5CN,
−78°C, 5 h, 64%, syn/anti=>98% de; (ii) CSA, Me2C(OMe)2, acetone, >95%

7515
The last step toward the target polyol unit 3 was addressed to the chiral oxazaborolidinone-
promoted aldol reaction accompanying the sequential asymmetric syn reduction with the same
oxazaborolidinone. It was known from our previous work that the coupled asymmetric
reactions proceed in one pot in the system using silyl enol ethers, not ketene silyl acetals.12
Indeed, the expected aldol reaction of segment A with silyl enol ether 16, prepared from segment
B in the usual manner, furnished the desired syn-1,3-diol 17 with >98% de in 64% yield, together
with a small amount of the corresponding aldol product (ca. 10%) (in Scheme 4). Acetalization
of 17 accomplished the effective synthesis of the protected compound of 3.13
The construction of the nine contiguous hydroxy stereocenters in the protected compound of
3 is now complete. Each selectivity obtained in the reaction sequence is presented in Fig. 2;
excellent to good enantioselectivities in the eight aldol reactions and one reduction were
achieved. Thus, our aldol strategy based on chiral oxazaborolidinone-promoted asymmetric
aldol reactions turned out to be quite reliable as a general way of constructing a variety of
iterative 1,3-polyacetate systems.
Figure 2. Each selectivity found in the reaction sequence
Acknowledgements
We thank the Ministry of Education, Science and Culture, Japan, for a Grant-in-Aid for
Scientific Research (Nos. 9554046 and 10640525).
References
1. (a) Oishi, T.; Nakata, T. Synthesis 1990, 635. (b) Rychonovsky, S. D. Chem. Rev. 1995, 95, 2021. (c) Schneider,
C. Angew. Chem., Int. Ed. Engl. 1998, 37, 1375. (d) Enders, D.; Hundertmark, T. Tetrahedron Lett. 1999, 40,
4169. (e) Garner, P.; Anderson, J. T. Org. Lett. 1999, 1, 1057. (f) Smith III, A. B.; Pitram, S. M. Org. Lett. 1999,
1, 2001. (g) Trieselmann, T.; Hoffmann, R. W. Org. Lett. 2000, 2, 1209.
2. (a) Kiyooka, S.-i. Rev. Heteroatom. Chem. 1997, 17, 245. (b) Kiyooka, S.-i.; Yamaguchi, T.; Maeda, H.; Kira, H.;
Hena, M. A.; Horiike, M. Tetrahedron Lett. 1997, 38, 3553. (c) Kiyooka, S.-i.; Madeda, H. Tetrahedron:
Asymmetry 1997, 8, 3371. (d) Kiyooka, S.-i.; Hena, M. A. J. Org. Chem. 1999, 64, 5511.
3. Isolation: (a) Whitfield, G. B.; Brock, T. D.; Ammann, A.; Gottlieb, D.; Carter, M. E. J. Am. Chem. Soc. 1955,
77, 4799. (b) Bergy, M. E.; Eble, T. E. Biochemistry 1968, 7, 653.
4. Structure Elucidation: (a) Rychnovsky, S. D.; Richardson, T. I. Angew. Chem., Int. Ed. Engl. 1995, 34, 1227. (b)
Richardson, T. I.; Rychonovsky, S. D. J. Org. Chem. 1996, 61, 4219.
5. Total synthesis: Richardson, T. I.; Rycohnovsky, S. D. J. Am. Chem. Soc. 1997, 119, 12360.
6. Kiyooka, S.-i.; Hena, M. A.; Goto, F. Tetrahedron: Asymmetry 1999, 10, 2871.
7. Kiyooka, S.-i.; Hena, M. A. Tetrahedron: Asymmetry 1996, 7, 2181.

7516
8. (a) Yamamoto, K.; Suzuki, S.; Tsuji, J. Chem. Lett. 1978, 649. (b) Sato, M.; Sunami, S.; Sugita, Y.; Kaneko, C.
Chem. Pharm. Bull. 1994, 42, 839. (c) Singer, R. A.; Carreira, E. M. J. Am. Chem. Soc. 1995, 117, 12360.
9. Chen, K.-M.; Hardtmann, G. E.; Prasad, K.; Repic, O.; Shapiro, M. J. Tetrahedron Lett. 1987, 28, 155.
10. Boaz, N. W. J. Org. Chem. 1992, 57, 4289.
11. Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini, G.; Dolling, U.-H.; Grabowski, E. J. J. Tetrahedron Lett.
1995, 36, 5461.
12. (a) Kaneko, Y.; Matsuo, T.; Kiyooka, S.-i. Tetrahedron Lett. 1994, 35, 4107. (b) Kiyooka, S.-i.; Kaneko, Y.;
Harada, Y.; Matsuo, T. Tetrahedron Lett. 1995, 36, 2821.
13. On analysis of 13C NMR correlations of 1,3-diol acetonide: see Rychonovsky, S. D.; Rogers, B.; Yang, G. J. Org.
Chem. 1993, 58, 3511. Selected physical data of the protected compound of 3: [h]2D3 −122.2 (c 0.20, CHCl3); IR
1
(neat) 2942, 1379, 1263, 1199, 1103 cm−1; H NMR (CDCl3, 400 MHz) l (ppm) 7.30–7.35 (m, 5H), 4.50 (s, 2H),
4.21 (d, 2H, J=11.95 Hz), 4.05–4.12 (m, 2H), 3.95–4.05 (m, 6H), 3.90 (dd. 2H, J=3.68, 12.0 Hz), 3.55 (dddd,
2H, J=5.70, 6.20, 8.81, 14.0 Hz), 2.05 (ddd, 1H, J=2.01, 6.25, 11.4 Hz), 1.99 (ddd, 1H, J=3.52, 7.31, 11.6 Hz),
1.83 (dd, 2H, J=6.43, 13.6 Hz), 1.77 (dd, 2H, J=6.21, 12.4 Hz), 1.60–1.63 (m, 2H), 1.42 (s, 3H), 1.41 (s, 6H),
1.37 (s, 3H), 1.36 (s, 3H), 1.34 (s, 3H), 1.33 (s, 6H), 1.25–1.55 (m, 13H), 1.18 (t, 2H, J=11.7), 0.90 (t, 3H,
J=6.44), 0.88 (s, 9H), 0.06 (s, 3H), 0.01 (s, 3H). 13C NMR (CDCl3, 100 MHz) l (ppm) 138.5, 128.4 (×2), 127.7
(×2), 127.6, 100.3, 98.3, 98.2, 98.1, 77.2, 73.1, 69.8, 66.6 (×2), 66.5, 65.6, 65.2, 63.6, 62.8, 59.0, 42.9, 42.2, 42.0,
41.6, 38.5, 36.4 (×2), 36.0, 32.4, 31.8, 30.3, 30.2, 29.5, 25.8 (×3), 25.4, 24.8, 24.7, 22.6, 20.0, 19.8, 18.8, 18.0, 14.1,
−4.2, −4.5. FABMS calcd for C48H84O10Si: 848.58 found: 848.87 [M+].
.