A total synthesis of (-)-antimycin A(3b)

Article abstract of DOI:10.1016/S0040-4039(00)01316-2

(-)-Antimycin A(3b), the antipode of natural antibiotic antimycin A(3b), was synthesized utilizing the asymmetric aza-Claisen rearrangement. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01316-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 7667–7671
A total synthesis of (−)-antimycin A3b
Tetsuto Tsunoda,* Takeshi Nishii, Makoto Yoshizuka, Chise Yamasaki, Tomonori Suzuki
and Shoˆ Itoˆ
Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Tokushima 770-8514, Japan
Received 3 July 2000; accepted 28 July 2000
Abstract
(−)-Antimycin A3b, the antipode of natural antibiotic antimycin A3b, was synthesized utilizing the
asymmetric aza-Claisen rearrangement. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: asymmetric synthesis; aza-Claisen rearrangements; amides; antibiotics; lactonization.
(+)-Antimycin A3 was isolated from streptomyces sp.1 as a component of antimycin A
complex (a mixture of A1–A8), which inhibited specifically the electron transfer activity of
ubiquinol-cytochrome c oxidoreductase.2 Among the components of antimycin A complex, A3,
which is available from Sigma Co., is one of the most active agents and has been widely used
in biological and biochemical investigations.3
Although the structure of antimycin A3 had been established as 3-methylbutanoate (+)-1 with
a unique nine-membered dilactone structure, recent analytical investigations using HPLC
disclosed that antimycin A3 was in fact a mixture of (+)-1, now renamed antimycin A3b, and its
acyloxy isomer at the 8-position, which is 2-methylbutanoate, named antimycin A3a.3b,4 Since the
separation of A3a and A3b is possible only with enormous effort, pure material of each
component is practically available for biological and biochemical investigations.5,6
Antimycin A3b (previous name A3) has been an attractive target of synthetic efforts and a
total synthesis7 and its formal syntheses8 have been accomplished. However, further effort to a
* Corresponding author. Fax: +81-88-655-3051; e-mail: tsunoda@ph.bunri-u.ac.jp
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01316-2

7668
practical and economical synthesis is still needed to supply sufficient amounts of pure antimycin
A3b for biological or biochemical studies.
We have developed the asymmetric aza-Claisen rearrangement,9 the thermal [3,3] sigmatropic
rearrangement of the enolate of carboxamides 2 to 3 (Scheme 1), and demonstrated its
applicability to straightforward short-step syntheses of natural products, including (−)-verrucari-
nolactone,10 D-allo-isoleucine,10 (−)-isoiridomyrmecin,11 and (+)-a-skytanthine.12 We applied the
methodology to the synthesis of (−)-antimycin A3b, the antipode of this naturally occurring
antibiotic, taking development of a practical synthetic route into consideration and having an
interest in biological activities.13 The result is reported herein.
Scheme 1.
(R)-(+)-Methylbenzylamine (4) was silylated with 1.0 equiv. of Me3SiOTf/DBU at 0°C. To the
resulting mixture were added [(CH3)2CH]3SiOTf (TIPSOTf, 1.0 equiv.) and acrolein (1.0 equiv.)
simultaneously to give the amine 5 (Scheme 2), in which the E-configuration of the olefinic
double bond was verified by NMR spectroscopy. Acylation of 5 afforded the amide 6, the
precursor of the proposed aza-Claisen rearrangement. The rearrangement was carried out under
standard conditions9 to give the amides as a four stereoisomeric mixture, from which the major
(7S,8R)-isomer 7a was obtained by SiO2 column chromatography as an inseparable 82:18
mixture with its (7R,8S)-isomer 7b (78% combined yield).
Scheme 2. (a) (i) TMSOTf, DBU, ether, 0°C, 1 h; (ii) TIPSOTf, acrolein, ether, 0°C–rt, 6 h, 64%; (b) hexanoyl
chloride, NEt3, CH2Cl2, 0°C, 98%; (c) (i) LHMDS, tol, −78°C; (ii) 120°C, 6 h, 78%; (d) I2, N,N-dimethyl-p-nitroani-
line, DME–H2O (3:2), rt, 3 days, 78%; (e) Bu3SnH, PhH, reflux, 91%; (f) (i) CsOH, t-BuOH, 0°C; (ii) prenyl bromide,
Py, ether, 0°C, 24 h, 40%
The iodolactonization14 of the mixture 7a,b afforded the iodo-lactone 8 (78%, enantiomeric
mixture derived from 7a,b) along with the stereoisomer 9 (4%), whose stereochemistry were

7669
determined by NOE experiments. The lactone 8 was reduced with Bu3SnH (2.4 equiv.) to 10.
After hydrolysis of 10 with 1 N CsOH aq. (1.0 equiv.) in t-BuOH, the solvent was evaporated
and the residue was dried in a desiccator (P2O5) under reduced pressure for 2 days. The resulting
gummy oil was treated with prenyl bromide (3.0 equiv.) and pyridine (3.0 equiv.) in ether at
0°C15 to yield the prenyl ester 11 (40%) along with the butenolide 12 (40%).
The Mitsunobu reaction16 of the hydroxy ester 11 with N,O-protected D-threonine 13 gave a
diastereomeric mixture of the diester 14a and its (7R,8R)-isomer 14b17 (Scheme 3). The TBDMS
and the prenyl groups on the mixture 14a,b were removed by acidic treatment, followed by a
palladium-catalyzed reaction, affording the seco acids 15a,b. The acids were converted into the
dilactones 17a,b in only 36% yield via the formation of the 2-pyridinethiol esters18 16a,b and
their treatment with AgClO4 at ambient temperature (the method of Gerlach)19. However, when
the reaction mixture was heated under reflux for 2 h the yield increased dramatically to 82%.
The diastereomer 17a and 17b were separated by SiO2 column chromatography.
Scheme 3. (g) 13, DEAD–PPh3, PhH, rt, 24 h, 100%; (h) 6N HCl, EtOH, rt, 24 h, 95%; (i) Pd(OAc)2, PPh3, NEt3,
HCOOH, dioxane, 100°C, 1 h, 85%; (j) 2,2%-dipyridyl disulfide, PPh3, PhH, rt, 3 h, 99%; (k) AgClO4, PhH, 80°C,
2 h, 82%; (l) TBAF, THF, 0°C, 5 min, 85%; (m) isovaleric anhydride, Py, 53%; (n) H2, Pd–C, AcOEt, 67%; (o) 20,
WSC, HOBt, NMM, DMF, rt, 95%; (p) H2, Pd–C, AcOEt, 89%
The O-isovaleryate 18 was obtained by quick treatment of 17a with Bu4NF at 0°C, followed
by acylation with isovaleric anhydride. The cbz group of 18 was removed by the hydrogenolysis
(Pd–C in ethyl acetate) to the amine 19, which was successfully acylated with 20, water-soluble
carbodiimide (WSC), 1-hydroxybenzotriazole hydrate (HOBt), and N-methylmolpholine
(NMM) in DMF to give the benzyl ether 21. Hydrogenolysis of 21 with Pd–C in ethyl acetate
led cleanly to (−)-antimycin A3b [(−)-1]20 in good yield, mp 183.5–184.0°C, [h]2D1 −74.2 (c 0.98,
CHCl3), whose physical properties compared well with those in the literature7b [mp 183–185°C,
[h]2D4 +80 (c 0.2, CHCl3)].

7670
Thus, the usefulness of the aza-Claisen rearrangement was amply demonstrated by the
efficient synthesis of (−)-antimycin A3b [(−)-1]. Further synthetic studies of not only (−)-1
analogs but also (+)-1 analogs and the investigation of their biological activities are now in
progress.
Acknowledgements
This work was supported partially by a Sunbor Grant from the Suntory Institute for
Bioorganic Research.
References
1. Dunshee, B. R.; Leben, C.; Keitt, G. W.; Strong, F. M. J. Am. Chem. Soc. 1949, 71, 2436–2437.
2. Wikstrom, M. K. F.; Berden, J. A. Biochim. Biophys. Acta 1972, 283, 403–420.
3. (a) Miyoshi, H.; Tokutake, N.; Imaeda, Y.; Akagi, T.; Iwamura, H. Biochim. Biophys. Acta 1995, 1229, 149–154.
(b) Barrow, C. J.; Oleynek, J. J.; Marinelli, V.; Sun, H. H.; Kaplita, P.; Sedlock, D. M.; Gillum, A. M.;
Chadwick, C. C.; Cooper, R. J. Antibiot. 1997, 50, 729–733. (c) Kim, H.; Esser, L.; Hossain, M. B.; Xia, D.; Yu,
C. A.; Rizo, J.; van der Helm, D.; Deisenhofer, J. J. Am. Chem. Soc. 1999, 121, 4902–4903, and references cited
therein.
4. (a) Abidi, S. L. J. Chromatogr. 1988, 447, 65–79. (b) Abidi, S. L. J. Chromatogr. 1989, 464, 453–458.
5. A NMR study in our laboratories disclosed that antimycin A3 purchased from Sigma Co. was a 6:4 mixture of
A3a and A3b.
6. Deisenhofer reported with surprise that the ‘pure’ Antimycin A1 purchased from Sigma Co. was in fact a mixture
of A1a and A1b. See Ref. 3c.
7. (a) Kinoshita, M.; Wada, M.; Aburaki, S.; Umezawa, S. J. Antibiot. 1971, 24, 724–726. (b) Kinoshita, M.;
Aburaki, S.; Wada, M.; Umezawa, S. Bull. Chem. Soc. Jpn. 1973, 46, 1279–1287.
8. (a) Aburaki, S.; Kinoshita, M. Bull. Chem. Soc. Jpn. 1979, 52, 198–203. (b) Nakata, T.; Fukui, M.; Oishi, T.
Tetrahedron Lett. 1983, 24, 2657–2660. (c) Wasserman, H. H.; Gambale, R. J. J. Am. Chem. Soc. 1985, 107,
1423–1424. (d) Wasserman, H. H.; Gambale, R. J. Tetrahedron 1992, 48, 7059–7070. (e) Inghart, T.; Frejd, T.
Tetrahedron 1991, 47, 6483–6492. (f) Kondo, H.; Oritani, T.; Kiyota, H. Heterocycles 2000, 49, in press.
9. Tsunoda, T.; Sakai, M.; Sasaki, O.; Sako, Y.; Hondo, Y. Tetrahedron Lett. 1992, 33, 1651–1653.
10. Tsunoda, T.; Tatsuki, S.; Shiraishi, Y.; Akasaka, M.; Itoˆ, S. Tetrahedron Lett. 1993, 34, 3297–3300.
11. Tsunoda, T.; Tatsuki, S.; Kataoka, K.; Itoˆ, S. Chem. Lett. 1994, 543–546. Itoˆ, S.; Tsunoda, T. Pure Appl. Chem.
1994, 66, 2071–2074.
12. Tsunoda, T.; Ozaki, F.; Shirakata, N.; Tamaoka, Y.; Yamamoto, H.; Itoˆ, S. Tetrahedron Lett. 1996, 37,
2463–2466.
13. (−)-Antimycin A3b had been synthesized once and had been compared with (+)-1 on inhibitory activity against
electron transport of rat liver mitochondria. See, Miyoshi, H.; Kondo, H.; Oritani, T.; Saitoh, I.; Iwamura, H.
FEBS Lett. 1991, 292, 61–63 and Ref. 8f.
14. An addition of N,N-dimethyl-p-nitroaniline completely prevented the deprotection of TIPS ether linkage of 8 on
Momose’s procedure. See, Takahata, H.; Uchida, Y.; Momose, T. J. Org. Chem. 1994, 59, 7201–7208.
15. This new procedure can be applied to open lactone rings (Tsunoda, T.; Nishii, T.; Suzuki, T.; Itoˆ, S., to be
published).
16. Mitsunobu, O. Synthesis 1981, 1–28.
17. 14b–17b derived from 7b were (7R,8R)-isomer of 14a–17a, respectively.
18. Endo, T.; Ikenaga, S.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1970, 43, 2632–2633.
19. Gerlach, H.; Thalmann, A. Helv. Chim. Acta 1974, 57, 2661–2663. In this paper, a benzene solution of the
pyridinthiol ester was treated with AgClO4 (1.0 equiv.) at ambient temperature. Kinoshita had also reported that
16c cyclized intramolecularly in a poor yield.7

7671
20. (−)-Antimycin A3b: colorless needles (rotamer mixture): mp 183.5–184.0°C (ether/pet. ether); [h]2D1 −74.2 (c 0.98,
1
CHCl3): H NMR (300 MHz, CDCl3) l 12.63 and 12.47 (total integr. 1H, s and br.s), 8.56 (1H, br.dd, J=8.1,
1.3 Hz), 8.52 (1H, d, J=1.8 Hz), 8.02 (1H, br.s), 7.25 (1H, dd, J=8.1, 1.3 Hz), 7.09 and 7.07 (total integr. 1H,
br.d, J=7.8 Hz and br.d, J=7.8 Hz), 6.92 and 6.90 (total integr. 1H, t, J=8.1 Hz and t, J=7.8 Hz), 5.76 (1H,
dq, J=7.8, 6.6 Hz), 5.32 and 5.29 (total integr. 1H, t, J=7.8 Hz and t, J=7.8 Hz), 5.12 and 5.16 (total integr.
1H, t, J=9.9 Hz and t, J=9.9 Hz), 4.98 (1H, dq, J=9.9, 6.3 Hz), 2.52 (1H, ddd, J=11.7, 9.9, 2.7 Hz), 2.26 (2H,
d, J=7.8 Hz), 2.15 (1H, septet d, J=7.8, 6.3 Hz), 1.78–1.63 (1H, m), 1.43–1.04 (5H, m), 1.32 (3H, d, J=6.6 Hz),
1.30 (3H, d, J=6.3 Hz), 0.99 (6H, d, J=6.3 Hz), 0.87 (3H, t, J=7.8 Hz); IR (neat) 3370, 1750, 1692, 1644, 1611
cm−1; MS (CI) m/z 520 (M+), 458, 418, 264, 236, 220, 202 (bp); HMRS m/z 520.2454 (520.24206 calcd for
C26H36N2O9).
.