Olefin cross-metathesis in the preparation of polycyclopropanes: Formal synthesis of FR-900848

Article abstract of DOI:10.1016/S0040-4039(00)01568-9

Olefin cross-metathesis was used for the efficient incorporation of an E-olefin between two cyclopropane residues. The generation of a bis-cyclopropyl alkene homodimer, followed by cross metathesis with a terminal vinyl polycyclopropane, provided excellent yields of the targeted compounds with good olefin stereoselectivity. The strategy was applied to generation of a quinta-cyclopropane intermediate found in a previous synthesis of the natural product FR-900848. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01568-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8723–8727
Olefin cross-metathesis in the preparation of
polycyclopropanes: formal synthesis of FR-900848
Christopher A. Verbicky and Charles K. Zercher*
Department of Chemistry, University of New Hampshire, Durham, NH 03824, USA
Received 7 March 2000; revised 2 September 2000; accepted 5 September 2000
Abstract
Olefin cross-metathesis was used for the efficient incorporation of an E-olefin between two cyclo-
propane residues. The generation of a bis-cyclopropyl alkene homodimer, followed by cross metathesis
with a terminal vinyl polycyclopropane, provided excellent yields of the targeted compounds with good
olefin stereoselectivity. The strategy was applied to generation of a quinta-cyclopropane intermediate
found in a previous synthesis of the natural product FR-900848. © 2000 Elsevier Science Ltd. All rights
reserved.
Keywords: olefin metathesis; cross-metathesis; FR-900848; synthesis; natural product; polycyclopropane.
The isolation and identification of two polycyclopropanated natural products has inspired
numerous synthetic efforts directed toward the assembly of these challenging molecular targets.
Both (−)-FR-900848 (1) and (−)-U-106305 (2) possess an unprecedented assembly of contiguous
cyclopropanes located on a fatty amide backbone (Fig. 1). The remarkable structural similarity
of these compounds stands in contrast to their divergent biological activities. (−)-FR-900848 (1)
was isolated from Streptovercillium fervens and was demonstrated to possess remarkably specific
biological activity with respect to its inhibition of non-filamentous fungi.1 In contrast to the
antifungal activity of 1, (−)-U-106305 was identified by Upjohn in a screen for inhibitors of
cholesteryl ester transfer protein (CEPT).2 Efforts to develop a clear structure–activity profile
will require the evolvement of efficient stereoselective approaches to these polycyclopropanated
skeletons. Most of the reported approaches toward these skeletons have focused on the
stereocontrolled formation of the contiguous cyclopropanes;3–7 however, the non-conjugated
(D14) E-alkene also presents its share of synthetic challenges. More specifically, the location of
the alkene between two cyclopropane units limits the options available for olefin formation.
* Corresponding author. Fax: (603) 862-4278; e-mail: ckz@christa.unh.edu
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01568-9

8724
Figure 1.
A number of approaches to the stereoselective formation of an E-alkene flanked by two
cyclopropane moieties have been described. These methods include Whitham elimination,8
hydroxyl-directed mono-cyclopropanation of an E,E-dienol,5 sulfone-modified Peterson olefina-
tion, followed by dissolving metal desulfurization,6 and a modified Julia coupling strategy.7 Poor
yields, additional synthetic steps for the removal of extraneous functionality, or modest olefin
selectivity plague many of these approaches; therefore, an alternative stereoselective approach to
the dicyclopropyl ethylene moiety is desired.
We contemplated the application of an olefin metathesis reaction.9 Olefin metathesis has
quickly become an attractive method for the assembly of complex functionality due to its power,
operational simplicity, and tolerance of a wide range of functional groups; however, application
of olefin metathesis to intermolecular cross-coupling reactions has been limited. Nevertheless, we
envisioned the application of the olefin cross-metathesis reaction reported by Grubbs10 to the
formation of the D14-E-alkene found in the two polycyclopropanated natural products. The
absence of reactive intermediates and the expectation of good E-selectivity were desirable
features of this method.
In order to test the compatibility of an olefin cross-metathesis strategy to the formation of
alkenes flanked by cyclopropanes, cinnamyl alcohol 3 was converted to the optically active
cyclopropane (89% ee)11 (Scheme 1) through the influence of a
D
-tartrate-derived dioxa-
boralane.12 Oxidation with TPAP/NMO13 and olefination with triphenylphosphonium methylide
provided vinyl cyclopropane 4. The exposure of 4 to Grubbs’ catalyst ((Cy3P)2Cl2RuꢀCHPh) in
refluxing methylene chloride provided the homodimer 5 (62%, 6:1; E:Z).14
Scheme 1. (a) Et2Zn, CH2I2,
D-tartrate-derived dioxaboralane, 98%; (b) TPAP, NMO; (c) Ph3PCH3Br, CH3Li, 83%
for two steps; (d) (Cy3P)2Cl2RuꢀCHPh (5 mol%), 62%

8725
The polycyclopropanated cross-metathesis partner was prepared in the fashion similar to that
previously reported by our group (Scheme 2).15 Reduction of butyne-1,4-diol 6 with lithium
aluminum hydride was followed by mono-protection with t-butyldimethylsilyl chloride (TBDM-
SCl) and triethylamine. The allylic alcohol was exposed to
D
-tartrate-derived dioxaboralane-
directed cyclopropanation conditions, which upon removal of the silyl protecting group
provided the trans-(1R,2R)-bis-hydroxymethylcyclopropane 7.15 Both alcohol functionalities
were oxidized and the dialdehyde chain-extended in two directions using a Horner–Emmons
reaction. Following reduction with excess DIBAL-H, the bis-allylic alcohol was exposed to
D
-tartrate-derived dioxaboralane-directed cyclopropanation conditions to provide bis-hydroxy-
methyl-tris-cyclopropane 8. Mono-protection of 8 with TBDMSCl and triethylamine, followed
by oxidation and chain extension provided an a,b-unsaturated ester, which was reduced with
excess DIBAL-H. Cyclopropanation of the allylic alcohol was once again directed by -tartrate-
D
derived dioxaboralane to provide tetrakis-cyclopropane 9. The removal of the protecting group
with tetra-n-butylammonium fluoride (TBAF) confirmed the anticipated C2-symmetry of bis-
hydroxymethyl-tetrakis-cyclopropane 10 (87% ee).15 Oxidation of 9, followed by olefination with
triphenylphosphonium methylide provided vinyl-tetrakis-cyclopropane 11 in high yield. The
vinyl-tetrakis-cyclopropane 11 and two equivalents of the homodimer 5 were combined and
exposed to Grubbs’ catalyst in refluxing methylene chloride for 20 hours. Following chromato-
graphic purification, the phenyl analog 12 of the polycyclopropanated side chain was formed in
excellent yield (94%)16 as a mixture of E- and Z-olefins (3.5:1).17
Scheme 2. (a) LiAlH4, 79%; (b) TBDMSCl, Et3N (1 equiv.), 86%; (c) Et2Zn, CH2I2,
(d) AcOH/THF/H2O (1:1:1), 86% for two steps; (e) TPAP, NMO; (f) (EtO)2P(O)CH2CO2Et, NaH, 78% for two steps;
(g) DIBAL-H, 73%; (h) Et2Zn, CH2I2, -tartrate-derived dioxaboralane, 60%; (i) TBDMSCl, Et3N (1 equiv.), 68%;
(j) TPAP, NMO; (k) (EtO)2P(O)CH2CO2Et, NaH, 92% two steps; (l) DIBAL-H, 53%; (m) Et2Zn, CH2I2, -tartrate-
D-tartrate-derived dioxaboralane;
D
D
derived dioxaboralane, 95%; (n) n-Bu4NF, 67%, 87% ee; (o) TPAP, NMO; (p) Ph3PCH3Br, CH3Li, 75% for two
steps; (q) 5 (two equiv.), (Cy3P)2Cl2RuꢀCHPh (5 mol%), 94%
Successful olefin cross-metathesis in the formation of 12 encouraged its application toward
FR-900848 (1). The preparation of an intermediate that would constitute a formal synthesis of
FR-900848 was envisioned (Scheme 3). Mono-protection of bis-hydroxymethylcyclopropane 7
with benzoyl chloride (BzCl), followed by oxidation and olefination provided vinyl cyclopropane

8726
13. The exposure of 13 to Grubbs’ catalyst provided homodimer 14 (11.3:1; E:Z). When 14 was
combined with vinyl-tetrakis-cyclopropane 11 and exposed to Grubbs’ catalyst, the cross-
coupled product 15 was formed in good yield (82%)16 with modest olefin (>5:1; E:Z) stereoselec-
tivity. Selective removal of the benzoyl protecting group with potassium hydroxide provided
polycyclopropane 16 (86% ee), which was identical to an intermediate in Barrett’s total
synthesis5c of (−)-FR-900848 (1).18
Scheme 3. (a) BzCl, Et3N, 35%; (b) TPAP, NMO; (c) Ph3PCH3Br, CH3Li, 71% for two steps; (d)
(Cy3P)2Cl2RuꢀCHPh (5 mol%), 64% (11.3:1, E:Z); (e) 11, (Cy3P)2Cl2RuꢀCHPh (5 mol%), 82%; (f) KOH, CH3OH,
89% (86% ee)
Preparation of 16 completed an enantioselective formal synthesis of (−)-FR-900848 and
demonstrated the utility of olefin cross-metathesis as a tool in the synthesis of polycyclo-
propanated fatty acids. Although the method is limited to the formation of E-alkenes, olefin
cross-metathesis is an attractive solution to dicyclopropyl alkene formation due to its opera-
tional simplicity and the good yields obtained in its application.
References
1. Yoshida, M.; Ezaki, M.; Hashimoto, M.; Yamashita, M.; Shigematsu, N.; Okuhara, M.; Kohsaka, M.;
Horikoshi, K. J. Antibiot. 1990, 43, 748–754.
2. Kuo, M. S.; Zielinski, R. J.; Cialdella, J. I.; Marschke, C. K.; Dupois, M. J.; Li, G. P.; Kloosterman, D. A.;
Spilman, C. H.; Marshall, V. P. J. Am. Chem. Soc. 1995, 117, 10629.
3. (a) Theberge, C. R.; Zercher, C. K. Tetrahedron Lett. 1994, 35, 9181. (b) Theberge, C. R.; Zercher, C. K.
Tetrahedron Lett. 1995, 36, 5495. (c) Cebula, R. E. J.; Hanna, M. R.; Theberge, C. R.; Verbicky, C. A.; Zercher,
C. K. Tetrahedron Lett. 1996, 37, 8341.
4. Taylor, R. E.; Ameriks, M. K.; LaMarche, M. J. Tetrahedron Lett. 1997, 38, 2057.
5. (a) Barrett, A. G. M.; Hamprecht, D.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 1996, 118, 7863. (b)
Barrett, A. G. M.; Doubleday, W. W.; Hamprecht, D.; Kasdorf, K.; Tustin, G. J. Pure Appl. Chem. 1997, 69, 383
and references cited therein. (c) Barrett, A. G. M.; Kasdorf, K. J. Am. Chem. Soc. 1996, 118, 11030.
6. Falck, J. R.; Mekonnnen, B.; Yu, J.; Lai, J.-Y. J. Am. Chem. Soc. 1996, 118, 6096.
7. Charette, A. B.; Lebel, H. J. Am. Chem. Soc. 1996, 118, 10327.
8. Barrett, A. G. M.; Kasdorf, K.; Williams, D. J. J. Chem. Soc., Chem. Commun. 1994, 1781.
9. After submission of this communication, a report on the metathesis coupling of fluorinated vinylcyclopropanes
was reported. Itoh, T.; Mitsukuru, K.; Ishida, N.; Uneyama, K. Org. Lett. 2000, 2, 1431.
10. O’Leary, D. J.; Blackwell, H. E.; Washenfelder, R. A.; Grubbs, R. H. Tetrahedron Lett. 1998, 39, 7427.
11. (a) Yasui, S. C.; Keiderling, T. A. J. Am. Chem. Soc. 1987, 109, 2311. (b) Theberge, C. R.; Verbicky, C. A.;
Zercher, C. K. J. Org. Chem. 1996, 61, 8792.
12. Charette, A. B.; Juteau, H. J. Am. Chem. Soc. 1994, 116, 2651.

8727
13. Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis 1994, 639.
1
14. Characterization data for new compounds: Compound 5: H NMR (360 MHz, CDCl3) l 7.42–7.0 (m, 10H), 5.28
(m, 2H), 1.98–1.80 (m, 2H), 1.74–1.60 (m, 2H), 1.30–1.17 (m, 2H), 1.16–1.10 (m, 2H); 13C NMR (90 MHz,
CDCl3) l 147.5, 131.0, 128.3, 125.7, 125.6, 26.6, 25.1, 16.7. HRMS (M+) m/z calcd for C20H20 260.1565, found
260.1559. Compound 9: [h]D=−107.0 (c 0.0242, CH2Cl2); 1H NMR (360 MHz, CDCl3) l 3.49–3.35 (m, 4H), 0.90
(s, 9H), 0.90 (m, 1H), 0.89–0.78 (m, 1H), 0.77–0.63 (m, 3H), 0.62–0.48 (m, 4H), 0.31–0.17 (m, 4H), 0.15–0.03 (m,
4H), 0.05 (s, 6H); 13C NMR (90 MHz, CDCl3) l 67.0, 66.7, 26.0, 19.7, 19.5, 18.6, 18.4 (2), 18.1 (2), 18.0, 8.3 (3),
1
8.25, 0.01, −5.1. Compound 11: H NMR (360 MHz, CDCl3) l 5.36 (ddd, 1H, J=17.1, 10.3, 9.7 Hz), 4.99 (d,
1H, J=17.1 Hz), 4.80 (d, 1H, J=10.3 Hz), 3.49–3.38 (m, 2H), 1.21–1.10 (m, 1H), 0.90 (s, 9H), 0.89–0.80 (m, 1H),
0.79–0.62 (m, 3H), 0.61–0.50 (m, 5H), 0.49–0.40 (m, 2H), 0.28–0.18 (m, 2H), 0.15–0.02 (m, 2H), 0.05 (s, 6H); 13C
NMR (90 MHz, CDCl3) l 141.9, 111.1, 66.7, 26.0, 22.3, 21.2, 19.5, 18.6, 18.4, 18.2 (2), 18.1, 11.8, 8.2, 8.1, 7.8,
0.01, −5.1. HRMS (MH+) m/z calcd for C21H37OSi 333.2614, found 333.2608. Compound 12: 1H NMR (360
MHz, CDCl3) l 7.40–7.03 (m, 5H), 5.17 (dd, 1H, J=15.2, 7.8 Hz), 5.08 (dd, J=15.2, 8.0 Hz), 3.55–3.35 (m, 2H),
1.98–1.68 (m, 1H), 1.66–1.54 (m, 1H), 1.21–1.00 (m, 2H), 0.90 (s, 9H), 0.95–0.31 (m, 11H), 0.30–0.18 (m, 1H),
0.18–0.02 (m, 4H), 0.05 (s, 6H); 13C NMR (90 MHz, CDCl3) l 132.5, 129.5, 128.3, 125.7, 125.5, 125.4, 66.7, 26.6,
26.0, 25.0, 22.0, 20.1, 19.5, 18.6, 18.4, 18.1 (2), 16.7, 11.6, 8.4, 8.2, 8.1, 7.8, −5.1. HRMS (MNH4+) m/z calcd for
C30H48NOSi 466.3505, found 466.3493. Compound 13: 1H NMR (360 MHz, CDCl3) l 8.15–8.00 (m, 2H),
7.61–7.52 (m, 1H), 7.50–7.41 (m, 2H), 5.45 (ddd, 1H, J=17.1, 10.1, 8.4 Hz), 5.09 (d, 1H, J=17.1 Hz), 4.91 (d,
1H, J=10.1 Hz), 4.26 (dd, 1H, J=11.5, 7.0 Hz), 4.19 (dd, 1H, J=11.5, 7.1 Hz), 1.57–1.47 (m, 1H), 1.45–1.30 (m,
1H), 0.95–0.74 (m, 2H); 13C NMR (90 MHz, CDCl3) l 166.7, 140.1, 132.9, 130.4, 129.6, 128.3, 112.8, 68.1, 20.9,
19.3, 12.0. Compound 14: [h]D=−13.77 (c 0094, CH2Cl2); 1H NMR (360 MHz, CDCl3) l 8.13–8.00 (m, 4H),
7.62–7.51 (m, 2H), 7.50–7.40 (m, 4H), 5.18 (m, 2H), 4.29–4.20 (m, 2H), 4.18–4.12 (m, 2H), 1.46–1.38 (m, 2H),
1.33–1.23 (m, 2H), 0.85–0.63 (m, 4H); 13C NMR (90 MHz, CDCl3) l 166.7, 132.9, 131.2, 130.5, 129.6, 128.3, 68.3,
19.8, 19.1, 11.9. HRMS (MNH+) m/z calcd for C24H28NO4 394.2018, found 394.2024. Compound 15: [h]D=
1
−109.38 (c 0.00256, CDCl3); H NMR (360 MHz, CDCl3) l 8.12–8.01 (m, 2H), 7.60–7.51 (m, 1H), 7.50–7.41 (m,
2H), 5.1 (m, 2H), 4.34–4.21 (m, 1H), 4.19–4.08 (m, 1H), 3.52–3.38 (m, 2H), 1.67–1.53 (m, 1H), 3.52–3.38 (m, 1H),
1.35–1.20 (m, 3H), 1.10–1.01 (m, 1H), 0.98–0.66 (m, 5H), 0.89 (s, 9H), 0.65–0.48 (m, 3H), 0.43–0.32 (m, 2H),
0.30–0.05 (m, 4H), 0.04 (s, 6H); 13C NMR (90 MHz, CDCl3) l 166.7, 132.8, 132.3, 130.5, 129.6, 128.9, 128.3,
68.4, 66.7, 26.0, 21.9, 20.9, 20.0, 19.8, 19.5, 19.0, 18.6, 18.4, 18.1 (2), 11.8, 11.5, 8.2, 8.1, 7.7, 0.01, −5.1. HRMS
(MNH+) m/z calcd for C32H47NO3Si 507.3295, found 507.3289.
15. McDonald, W. S.; Verbicky, C. A.; Zercher, C. K. J. Org. Chem. 1997, 62, 1215.
16. We can offer no rationalization for this observation that the isolated yield appears to be higher than predicted
by simple statistical analysis.
17. Separation of diastereomers was attempted by column chromatography, although removal of all diastereomeric
side products was not possible.
1
18. Assignment made by comparison to optical rotation and H and 13C-characterization data reported in Ref. 5c.
.